---
title: 12.09 The Vendor Risk Management Extra
description: Activate the Vendor Risk Management Extra on top of SimpleRisk's own vendor list, grant its permissions alongside the Vendor Management ones, set up risk…
---

[Skip to content](https://support.simplerisk.com/kb/12-09-the-vendor-risk-management-extra#main-content)

English

Show submenu for translations

[Customer portal](https://support.simplerisk.com/tickets?hsLang=en)

[![SimpleRisk logo of a man walking a tight rope](https://support.simplerisk.com/hs-fs/hubfs/simplerisk_logo_long_small-4.png?width=377&height=72&name=simplerisk_logo_long_small-4.png)](https://www.simplerisk.com/)

Open main navigation

Close main navigation

- English
  
  Show submenu for translations
- [Customer portal](https://support.simplerisk.com/tickets)
- [Contact us](https://www.simplerisk.com/about-us/contact-us)

[Contact us](https://www.simplerisk.com/about-us/contact-us)

 How can we help you?

- There are no suggestions because the search field is empty.

1. [SimpleRisk Knowledge Base](https://support.simplerisk.com/kb?hsLang=en)
2. [Administrator Guide](https://support.simplerisk.com/kb/administrator-guide?hsLang=en)
3. [12 Operations](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#12-operations)

# 12.09 The Vendor Risk Management Extra

## Activate the Vendor Risk Management Extra on top of SimpleRisk's own vendor list, grant its permissions alongside the Vendor Management ones, set up risk tiers, the intake questions that place a vendor in one and the tier policies and reminders that drive due diligence, keep the evidence type and contract term catalogs and the file size limit, set how findings are scored when they become risks and what the offboarding checklist holds, choose the SCF controls each tier requires, set the budgets and hourly limit for AI contract term extraction and know what it sends to the AI provider, import vendors from Assessments contacts, and understand the Overview and Reports pages and their limits, fourth-party dependencies, what happens to control coverage when the SCF Extra changes, team visibility, encryption of vendor details, the API and what deactivation deletes.

> **Requires:** *Vendor Risk Management Extra*
> 
> SimpleRisk itself keeps a vendor list, **Manage Vendors**, with each vendor's profile, status, teams and links to assets and risks. This Extra adds risk tiers, intake questionnaires, services and review dates, contacts, a computed risk score, due diligence, evidence, contracts, findings, fourth-party dependencies, an offboarding checklist, and the **Vendor Risk Dashboard** and six vendor reports to the same vendors. With the Secure Controls Framework (SCF) Extra active as well, it also records each vendor's coverage of the SCF controls its tier requires. Without it, those parts of the menu and the vendor page show as locked.

## Why this matters

Third-party risk usually starts as a spreadsheet: vendor name, a contact, a note about what they do for you. SimpleRisk already moves the name, the owner and the links to your assets and risks into **Manage Vendors**. What it doesn't do on its own is tell you how much each vendor matters. That's this Extra's job.

The Extra adds the parts of a vendor risk program that sit on top of the list. A short intake questionnaire scores a vendor's inherent risk and proposes a tier. An approver confirms the tier, and the tier sets how often the vendor gets a fresh look. Each vendor also gains a services description, a next review date, its contacts and a computed risk score with a history of why it changed, and the step that makes a vendor Active now needs an approver and a tier. Due diligence closes the loop: analysts record reviews on the vendor or send it a questionnaire through the Assessments Extra, completing a review schedules the next one from the tier's interval, and the vendor's business owner is emailed when a review comes due. Evidence and contracts keep the paper behind it: certificates and reports with their expiry dates, the evidence each tier must have on file, and contracts with their dates and a catalog of key terms compared with your preferred positions. Findings track what the reviews turned up, each with an owner and a due date, and a serious one becomes a normal risk in the risk register. When the relationship ends, an offboarding checklist stands between Offboarding and Terminated. Around all of that, each vendor records the other vendors it relies on, the **Vendor Risk Dashboard** shows where to look first, the vendor reports on the Reports page answer the standing questions and export the register, and, with the SCF Extra, each tier names the Secure Controls Framework (SCF) controls its vendors must meet, which analysts then record vendor by vendor.

It can also do the slow part of contract tracking for you. With an AI provider configured in SimpleRisk, a user with the extraction permission has the provider read a contract's PDF or Word document and propose a value for each contract term and date, plus the contract's total value, currency and SLA summary, each with the passage it came from. The proposals wait for someone who manages contracts to accept, edit or reject them; nothing is written to a contract automatically. Step 17 covers the settings and exactly what leaves your instance.

## Before you start

- **Administrator access.** Activating and deactivating the Extra are done by an administrator from the Settings Hub. Everything else on the Extra's settings page (tiers with their templates and required evidence, intake questions, notifications, catalogs and the other sections) is open to an administrator and to any user who holds **Able to Configure Vendor Risk Management**.
- **The Vendor Management permissions in place.** Every Extra screen sits inside **Manage Vendors**, so every user of the Extra also needs **Allow Access to "Vendor Risk" Menu**. Step 2 explains how the two permission groups combine.
- **A plan for who sees what.** Vendor records carry contact details, so decide which roles get which permissions before you turn the Extra on.
- **An encryption key, once built-in encryption is available in your release, if you want people to save contact details, evidence, contracts, finding descriptions, offboarding notes, or notes on a dependency or a control assessment.** Activation itself does not need one. See "Vendor details and encryption" below.
- **The Assessments Extra, only if you want to import vendors from assessment contacts, send due diligence questionnaires, or add evidence from the files attached to their answers.** The register, due diligence recorded by hand, evidence and contracts work fine without it. SimpleRisk's messages call it the Risk Assessment Extra.
- **The queue cron jobs running.** Review, evidence, contract and finding reminders, automatic reassessment and the questionnaire status check run as queued background jobs, so `cron_queue_loader.php` and `cron_queue_worker.php` must be scheduled. See [The Cron Jobs](https://support.simplerisk.com/kb/02-07-the-cron-jobs?hsLang=en). Restart the long-running workers after you upgrade, so they pick up the Extra's new jobs. Contract term extraction runs on the queue worker too, and a worker started before the upgrade can fail every extraction with the provider error, or leave it waiting, until it's restarted.
- **Outgoing email configured,** for the review, evidence, contract and finding reminders and for questionnaires. See [Email and the Notification Extra](https://support.simplerisk.com/kb/07-02-email-and-the-notification-extra?hsLang=en).
- **The Team-Based Separation Extra, only if you want to limit vendors by team.** Without it, everyone who can open Manage Vendors sees every vendor.
- **Governance access, only if you want control coverage.** Required controls, the vendor **Controls** tab and the **Vendor Control Coverage** report use SimpleRisk's controls (any framework, SCF or your own), so everyone who uses them also needs **Allow Access to "Governance" Menu**. Everything else works without it. Step 14 explains.

## Step-by-step

### 1. Activate the Extra

1. Open the Settings Hub from the Settings cog at the top right of any page.
2. Choose the **Extras** chip, then the **Vendor Risk Management Extra** tile.
3. Select **Activate**.

Activation creates the Extra's tables, adds its permissions to the **Vendor Management** group on the permission screens, and loads four risk tiers, six intake questions, six evidence types, sixteen contract terms and three offboarding checklist items. The evidence types, terms and checklist items are loaded only into an empty catalog, so reactivating never overwrites your edits. Administrators are granted the new permissions automatically. Everyone else gets nothing until you grant it. Your existing vendors, their links and their tags are untouched; they start with no tier.

The Extra's page then shows **ACTIVATED** with its version and the **Deactivate** button (administrators only), a note pointing to **Add and Remove Values** for the vendor category and data type lists, and a side navigation that lists every settings section by name. Choose a section to open it on its own, with the navigation staying in view. The sections are grouped as **Requirements** (**Evidence types** and **Contract terms**, the things a tier can require, then **Vendor tiers**, where each tier's review interval, questionnaire template, required evidence and required controls are set: steps 5, 8, 9, 10 and 14), **Due diligence** (**Intake questions**, step 6, and **Notifications**, step 8), **Findings** (**Finding settings** and **Offboarding checklist**, steps 12 and 13) and **AI** (**Contract term extraction**, step 17, shown only when that AI capability is switched on). A group appears only when you may configure what is in it. The address of a section ends in `#vr-section-` followed by its name, for example `#vr-section-offboarding-checklist`, so you can link straight to one.

In the left menu, **Vendor Risk** now holds, in this order, **Manage Vendors**, **Onboard Vendors**, **Conduct Due Diligence** for users who can run due diligence, **Manage Contracts** for users who can read evidence or contracts, **Manage Findings** for users who can read findings. Neither the overview nor the reports are menu entries. The **Vendor Risk Dashboard** is under Reporting, then Dashboards, and the six vendor reports (**Vendor Register**, **Overdue Vendor Reassessments**, **Expiring Vendor Evidence and Contracts**, **Vendor Control Coverage**, **Fourth-Party Concentration** and **Vendor Risk Heat Map**) are under Reporting, then Reports, all under the **Vendor Risk** filter. They appear only while the Extra is active, and each report is offered only to users who hold the permission to read it (step 15). A vendor's page gains the **Findings** tab for the same users, a **Dependencies** tab and an **Offboarding** tab for everyone who can view the Extra's data, and a **Controls** tab for users who can run due diligence and also hold **Allow Access to "Governance" Menu**.

### 2. Grant the permissions

The permissions are listed in one group, **Vendor Management**: SimpleRisk's own rows, plus the Extra's rows while it is active. A user of the Extra needs something from both sets. Deactivating the Extra removes its rows and leaves SimpleRisk's.

**Vendor Management** is part of SimpleRisk and owns the vendor record:

#### `vendor`

- **Display name:** Allow Access to "Vendor Risk" Menu
- **What it allows:** The **Vendor Risk** menu and **Manage Vendors**, read-only, plus the **Vendors** field on assets and risks. Required for every Extra permission below to do anything.

#### `vendor_add`

- **Display name:** Able to Add Vendors
- **What it allows:** Adding vendors, including from an intake's **New vendor name** and from the Assessments import

#### `vendor_edit`

- **Display name:** Able to Edit Vendors
- **What it allows:** Editing a vendor's profile, changing its status and saving its linked assets (which also needs **Able to Edit Assets**)

#### `vendor_delete`

- **Display name:** Able to Delete Vendors
- **What it allows:** Deleting vendors, which also deletes their Extra data

The rows the Extra adds to **Vendor Management** own its data:

#### `vr_view`

- **Display name:** Able to View Vendors
- **What it allows:** Shows the Extra's columns, panels and tabs on **Manage Vendors** (tier, tier history, services, review date, contacts, fourth-party dependencies) and opens **Onboard Vendors** with the intake list and each intake's detail, the **Vendor Risk Dashboard** and **Vendor Reports**, each showing only the sections the user's other permissions allow. Read-only.

#### `vr_manage`

- **Display name:** Able to Manage Vendors
- **What it allows:** Everything `vr_view` does, plus editing services and the next review date, adding, editing and removing contacts, adding, editing and deleting fourth-party dependencies, and submitting intakes, so this is the requester's permission. It also works a vendor's offboarding checklist: ticking steps, writing their notes, and **Start checklist** for a vendor in Offboarding that has none. It does not create, edit or delete the vendor itself.

#### `vr_approve`

- **Display name:** Able to Approve Vendors
- **What it allows:** Deciding intakes (approve, reject, or approve into a different tier) and setting a vendor's tier directly. Together with `vendor_edit`, moving a vendor to Active from the **Change Status** card, and terminating a vendor at the end of offboarding, from the card or the **Offboarding** tab. Needs `vr_view`, or a permission that includes it, to open anything.

#### `vr_configure`

- **Display name:** Able to Configure Vendor Risk Management
- **What it allows:** Implies the ability to view vendors. Lets a user change the tiers, the intake questions, the tier policies, the notification and file size settings, the evidence types, the contract terms, the finding settings and the offboarding checklist items, and, together with **Allow Access to "Governance" Menu**, the controls each tier requires (step 14). Together with `vr_manage` and the Assessments permission, it runs the import described in step 4. Like `vr_approve`, it reveals the points behind each answer on an intake. It does not let anyone read a vendor's evidence, contracts or findings.

#### `vr_due_diligence`

- **Display name:** Able to Run Vendor Due Diligence
- **What it allows:** Together with `vr_view` (or a permission that includes it), the vendor's **Assessments** tab and the **Conduct Due Diligence** page: recording, editing and deleting assessment records and refreshing a questionnaire's status. Sending a questionnaire also needs **Allow Access to "Assessments" Menu**, **Able to Add Questionnaires** and **Able to Send Questionnaires**. It also reads evidence: the vendor's **Evidence** tab, read-only, and the **Manage Contracts** page with evidence dates on its expiry calendar and nothing else. And it reads findings: the vendor's **Findings** tab and the **Manage Findings** page, read-only, with **Create finding** on each record for a user who also holds `vr_findings_manage`. Together with **Allow Access to "Governance" Menu**, it opens the vendor's **Controls** tab, records and clears control assessments, unlinks waivers and reads the **Vendor Control Coverage** report; linking a waiver also needs **Able to View Exceptions**. On its own it opens nothing.

#### `vr_evidence_manage`

- **Display name:** Able to Manage Vendor Evidence
- **What it allows:** Together with `vr_view` (or a permission that includes it), the vendor's **Evidence** tab: adding, editing, replacing and deleting evidence items and their files, and the evidence dates on the **Manage Contracts** calendar. Adding evidence from a questionnaire answer also needs **Allow Access to "Assessments" Menu** and **Able to View Questionnaire Results**.

#### `vr_contracts_view`

- **Display name:** Able to View Vendor Contracts
- **What it allows:** Together with `vr_view` (or a permission that includes it), the vendor's **Contracts** tab, read-only, with each contract's value, SLA summary, notes, document and terms, and the whole **Manage Contracts** page: the contract dates on its calendar, the contract list and the terms matrix.

#### `vr_contracts_manage`

- **Display name:** Able to Manage Vendor Contracts
- **What it allows:** Everything `vr_contracts_view` does, plus adding, editing and deleting contracts and setting their term values. It also reads a contract's AI term proposals and accepts, edits or rejects them.

#### `vr_contracts_extract`

- **Display name:** Able to Extract Terms from Vendor Contracts
- **What it allows:** Together with `vr_view` (or a permission that includes it) and `vr_contracts_view` (or `vr_contracts_manage`), starting and cancelling an AI extraction of a contract's terms and reading its proposals. It sends the contract document's text and your term catalog to the AI provider (step 17). On its own it opens nothing, and it never lets anyone read or change a contract.

#### `vr_findings_manage`

- **Display name:** Able to Manage Vendor Findings
- **What it allows:** Together with `vr_view` (or a permission that includes it), the vendor's **Findings** tab and the **Manage Findings** page: adding, editing and deleting findings and unlinking them from a risk. Creating a risk from a finding also needs **Able to Submit New Risks** and **Allow Access to "Risk Management" Menu**; linking a finding to an existing risk needs **Allow Access to "Risk Management" Menu** and **Able to Modify Risk Details**, and a risk the user can see.

The way they combine is simple and strict: an Extra permission counts only for a user who also holds **Allow Access to "Vendor Risk" Menu**. A user with **Able to View Vendors** but not the menu permission sees no **Vendor Risk** menu and gets a 403 from every Extra route. Permissions are cached in a user's session when they sign in, so people who are already signed in need to sign out and back in before a new grant takes effect.

Nine points are easy to miss.

- **Able to Manage Vendors** is not the same as the **Manage Vendors** page. It manages the Extra's data about a vendor. Creating, editing and deleting the vendor itself takes the Vendor Management permissions. A requester who adds new vendors through intakes needs **Able to Add Vendors** too.
- **Able to Approve Vendors** does not open any page on its own. Every approver needs **Able to View Vendors** at least. That's enough to decide intakes and set tiers, because approving an intake makes the vendor Active without **Able to Edit Vendors**. To use the **Change Status** card, the approver needs **Able to Edit Vendors** as well.
- The tier and intake question editors, and the other configuration cards, sit on the Extra's own page, which only administrators can open, and they appear only for a user who holds **Able to Configure Vendor Risk Management**. Administrators get it when the Extra is activated. For everyone else the permission matters for the API and the import.
- **Able to Run Vendor Due Diligence** behaves like **Able to Approve Vendors**: without **Able to View Vendors**, or a permission that includes it, the user sees no **Assessments** tab and no **Conduct Due Diligence** entry, and the routes answer 403. Sending a questionnaire from the vendor takes **Allow Access to "Assessments" Menu**, **Able to Add Questionnaires** and **Able to Send Questionnaires** as well, the same permissions the Assessments Extra asks for when someone adds and sends a questionnaire there, because the send creates one in that Extra and emails it. Without all three the user can still record reviews by hand. The tab's **Questionnaire Results** link needs that permission plus **Able to View Questionnaire Results**, with the Assessments Extra active.
- **Evidence and contracts are separate families, on purpose.** The people who collect SOC 2 reports and the people who negotiate liability caps are usually different, and the commercial terms are often more sensitive than the security paperwork. No evidence permission shows a contract, and no contract permission shows evidence or due diligence. The one bridge runs the other way and stops at evidence: **Able to Run Vendor Due Diligence** reads evidence (but never changes it), so analysts can see the certificates behind the reviews they record. Those analysts also get **Manage Contracts** in their menu, showing only the evidence dates on its calendar. The name looks odd for them, and it's worth telling them so before they ask.
- **Findings sit between the Extra and the risk register.** Reading findings takes **Able to Run Vendor Due Diligence** or **Able to Manage Vendor Findings**; nothing else, contract and evidence permissions included, shows them. Turning a finding into a risk takes the Core permissions a person would need to submit that risk by hand, **Able to Submit New Risks** and **Allow Access to "Risk Management" Menu**, and linking an existing one takes the permissions to edit that risk's vendors, **Allow Access to "Risk Management" Menu** and **Able to Modify Risk Details**. Without those, a findings manager can record and close findings but sees no **Create risk** or **Link existing risk**. A linked risk's ID, status and link are shown only to users with **Allow Access to "Risk Management" Menu** who can see the risk; everyone else sees **Linked to a risk**. A findings manager doesn't need any vendor-record permission to assign owners: **Able to Manage Vendor Findings** brings the **Owner** list, every enabled user they can see, and **Able to Manage Vendor Contracts** does the same for a contract's owner. That list shows those people's names to anyone holding either permission.
- **Control coverage leans on SimpleRisk's Governance permissions.** Controls, frameworks and exceptions are Governance data, so the Extra asks for the same permissions SimpleRisk does. The vendor's **Controls** tab, recording an assessment, clearing one, unlinking a waiver and the **Vendor Control Coverage** report need **Able to Run Vendor Due Diligence** plus **Allow Access to "Governance" Menu**. Linking a waiver also needs **Able to View Exceptions**, because the picker shows exception names; unlinking doesn't, because it reveals nothing. Without **Able to View Exceptions**, a waived control reads **Waived** or **Waiver pending approval** with no exception name. Choosing a tier's required controls needs **Able to Configure Vendor Risk Management** plus **Allow Access to "Governance" Menu**, so a configurer without the Governance permission keeps every other card but not that one. There's no locked Controls tab: a user who lacks one of these simply doesn't see it.
- **The dashboard and Reports show each user only their own domains.** Both open for anyone with **Able to View Vendors**, but every tile, report and register column is there only when the user can read its source: review figures for due diligence, evidence figures for evidence readers, contract figures for contract readers, finding figures for findings readers, asset figures for **Allow Access to "Asset Management" Menu**, and coverage for the coverage permissions above while the SCF Extra is available. A section the user can't read is left out entirely, never shown as zero, and the warning signs that rank **Vendors that need attention** follow the same rule, so a contracts-only user's list can't move when a finding changes. The register export follows it column group by column group.
- Keep requesters and approvers apart by role. Give the people who submit intakes **Able to Manage Vendors** (plus **Able to Add Vendors** if they bring in new vendors), give the people who decide them **Able to Approve Vendors** plus **Able to View Vendors**, and put both on one person only when you have to.

**Able to Extract Terms from Vendor Contracts** is the one to think about before you grant it, because it's the permission that sends contract text off your instance. It needs **Able to View Vendor Contracts** as well, so it never widens who can read a contract. Three combinations cover most teams:

#### **Able to View Vendor Contracts** only

- **Extract terms:** No
- **Read proposals:** No
- **Accept, edit, reject:** No

#### **Able to View Vendor Contracts** + **Able to Extract Terms from Vendor Contracts**

- **Extract terms:** Yes
- **Read proposals:** Yes
- **Accept, edit, reject:** No

#### **Able to Manage Vendor Contracts**

- **Extract terms:** No
- **Read proposals:** Yes
- **Accept, edit, reject:** Yes

#### **Able to Manage Vendor Contracts** + **Able to Extract Terms from Vendor Contracts**

- **Extract terms:** Yes
- **Read proposals:** Yes
- **Accept, edit, reject:** Yes

A contracts viewer without either of the other two never sees unconfirmed AI output. Administrators hold the extraction permission automatically, like every permission; there's no separate on/off switch for the feature, so the permission and the provider configuration together are the switch. If you don't want contract text sent to a provider at all, leave the permission off every role and keep in mind that administrators can still extract.

### 3. Categories and data types live in Add and Remove Values

The vendor category and data type lists belong to SimpleRisk, not to the Extra. Edit them under Settings cog → Settings Hub → **Add and Remove Values**, where they appear as **Vendor Categories** and **Vendor Data Types**. The Extra's page links there. Like the other lists on that page, they're for administrators only, and **Able to Configure Vendor Risk Management** doesn't change that.

A new installation starts with these values:

- **Vendor Categories:** Software / SaaS; Infrastructure / Cloud; Professional Services; Payment / Financial; Data Processing; Facilities / Physical; Other
- **Vendor Data Types:** Personal data; Payment card data; Health data; Credentials / secrets; Confidential business data; Public data only

Renaming is the safe operation: vendors keep pointing at the item and the new name shows everywhere at once. Deleting an item is allowed even when vendors use it. Deleting a category clears it from those vendors, and deleting a data type removes it from them, and SimpleRisk writes an audit entry for each vendor it changes. Reassign vendors first if you care where they end up.

### 4. Import vendors from Assessments contacts

If the Assessments Extra is active, your assessment contacts probably already name most of your vendors, because each contact carries a company. The import turns those companies into vendors without retyping anything.

The user running the import needs **Able to Configure Vendor Risk Management**, **Able to Manage Vendors** and **Allow Access to "Assessments" Menu**, as well as **Allow Access to "Vendor Risk" Menu**. The Assessments Extra must be active. When any of those is missing, the **Import contacts from Assessments** button doesn't appear on **Manage Vendors**.

1. Open **Vendor Risk**, then **Manage Vendors**, from the left menu and select **Import contacts from Assessments**.
2. The dialog groups the assessment contacts that aren't linked to a vendor yet by company. Tick the companies you want, or tick **Select All** at the top to take every company in the list.
3. Select **Import selected**.

For each company you pick, SimpleRisk creates a vendor with that name unless one already exists, then links each of the company's contacts to it as an **Assessment Contact**. A new vendor starts as a Prospect with no category or team. Creating it needs **Able to Add Vendors**. Without that permission the import only links contacts to vendors that already exist and the user can see, and it refuses the rest with *You do not have permission to add vendors. Choose an existing vendor, or ask an administrator for the Able to Add Vendors permission.*

Team separation adds one more catch: a non-administrator can't create a vendor without a team, so for them the import is refused for any company that doesn't have a vendor yet, and an administrator needs to run it. Running the import again for the same contacts changes nothing, and the assessment contacts themselves are never modified. The dialog shows up to 200 companies and 500 contacts at a time; when more exist it says so, and you open it again after importing the first batch.

Matching is by vendor name, ignoring case. If a vendor with that name exists but is in a team the importing user can't see, the import is refused as if the vendor didn't exist.

### 5. Set up the risk tiers

A tier is a named band of inherent risk with a review interval. On the Extra's page, the **Vendor tiers** card holds the ladder. Each row has a **Tier name**, a **Minimum score (0-100)**, a **Review interval (days)** and a read-only **Vendors in tier** count.

When you activate the Extra, it loads four tiers:

#### Critical

- **Minimum score:** 75
- **Review interval:** 180 days

#### High

- **Minimum score:** 50
- **Review interval:** 365 days

#### Medium

- **Minimum score:** 25
- **Review interval:** 730 days

#### Low

- **Minimum score:** 0
- **Review interval:** 0 (no scheduled review)

A scored intake lands in the tier with the highest minimum score that the intake's score reaches. A score of 60 lands in High, a score of 75 in Critical, and a score of 10 in Low. Select **Add tier** at the top right of the section to add a tier, the pencil on a row to edit it, and the trash can on a row to delete it (SimpleRisk asks you to confirm first). Adding and editing open a dialog with **Save** and **Cancel**. Each change is saved as soon as you confirm it, and the whole ladder is checked together, so a change that breaks a rule below is refused and nothing changes.

A change is refused unless the resulting ladder follows these rules:

- **2 to 8 tiers.**
- **Exactly one tier at a minimum score of 0.** That tier catches everything below the next threshold, so every score has a home.
- **Unique names and unique minimum scores.** Names are compared ignoring case and can be up to 100 characters. Scores are whole numbers from 0 to 100.
- **A review interval from 0 to 3650 days.** Zero means no scheduled review.

A tier that vendors hold is deleted by moving them. The row says how many vendors are assigned to it. When you select its trash can, SimpleRisk asks which tier to move them to, then moves every vendor holding the tier (terminated vendors too) and deletes it in one step. Each move is recorded in the vendor's tier history as a tier set by an analyst, with a note that the tier was deleted, and a vendor that has no next review date gets the new tier's interval (a date a vendor already has is never moved). Under team separation the count covers only the vendors you can see, but every vendor holding the tier is moved. The tier's policy, required evidence, required terms and required controls are deleted with it, and a delete that would leave a ladder that breaks the rules above is refused. Under team separation the count covers only the vendors you can see, but removal is refused while any vendor holds the tier, so a tier showing 0 can still be refused, and the message gives the full count. Renaming a tier and changing its threshold or interval are always allowed.

If another administrator changes the ladder while you have the page open, your change is refused with *The tiers were changed by someone else since you opened this page. Reload them and make your changes again.* The card reloads the current ladder. Make your edits again and save.

**Threshold changes affect future intakes only.** Vendors keep the tier they already have, and an intake scored last month keeps its score and its proposed tier. If you raise Critical from 75 to 80, nobody is demoted. The next intake that scores 77 lands in High. A vendor's tier history shows tier names as they were when each change happened.

The review interval drives the due diligence cycle. When a vendor receives a tier and its **Next Review Date** is empty, SimpleRisk fills the date in with today plus the interval. A date that is already set is not moved by a tier change, and a tier with an interval of 0 sets nothing. When a review of the vendor is completed, the date moves to the review date plus the interval, but never backward. The vendor's business owner is reminded as the date approaches; step 8 sets that up.

### 6. Edit the intake questions

The **Intake questions** card, below the tiers, lists the questions a requester answers. Each row shows the question text, its weight, how many answer options it has and an **Enabled** switch. **Add question** opens a form, and **Edit** and **Delete** act on a row.

A fresh activation loads six questions, each with four answers worth 0, 1, 2 and 3 points:

- **How sensitive is the data this vendor will handle?:** 3
- **How many records will the vendor hold or process?:** 2
- **What access will the vendor have to our systems or network?:** 3
- **How badly would an outage at this vendor disrupt our operations?:** 3
- **Is the vendor part of a customer-facing or regulated service?:** 2
- **How much does the vendor rely on subprocessors?:** 1

In the question form you set the **Question text** (up to 500 characters), the **Weight (1-10)**, whether the question is **Enabled**, and the **Answer options**. A question needs between 2 and 10 options. Each has an **Answer** label of up to 255 characters, unique within the question, and **Points (0-10)**. At least one option must be worth more than 0 points. A question whose options are all 0 can't change any score, so the form refuses it. A disabled question isn't asked. Deleting a question asks for confirmation and can't be undone, but it never touches the past: every intake keeps a copy of the questions and answers as they were when it was submitted.

Like the tiers, changes to questions, weights and points affect future intakes only. SimpleRisk never rescores an old intake.

If no tier exists, or no enabled question has at least two answers, requesters see a message that an administrator must set up tiers, or enable a question, before an intake can be submitted. A question with fewer than two answer options can't be answered, so it blocks submission until you fix it or disable it.

#### How the score is computed

Each answer earns the points of the option chosen. A question's best possible points are the highest points among its options. The intake score weights each answer by its question's weight, adds the results up, and expresses the total as a percentage of the best possible total, rounded to the nearest whole number:

score = round(100 × Σ(weight × points) ÷ Σ(weight × best points))

Take an intake with three enabled questions. Data sensitivity has weight 3 and the requester picks an answer worth 2 of a possible 3. Network access has weight 3 and the answer is worth 3 of 3. Subprocessors has weight 1 and the answer is worth 0 of 3. The numerator is 3 × 2 + 3 × 3 + 1 × 0 = 15. The best possible total is 3 × 3 + 3 × 3 + 1 × 3 = 21. The score is round(100 × 15 ÷ 21), which is 71, so with the default ladder the proposed tier is High, just under Critical's 75.

Because the score is a share of the best possible total, adding a question or changing a weight shifts every future score a little. If you tune the questions, check the outcome against a few vendors whose tier you already know before you let requesters loose on them.

### 7. Decide who submits and who decides

Intakes run on **Onboard Vendors**. The requester submits with **Able to Manage Vendors**, plus **Able to Add Vendors** to bring in a vendor that isn't in the list yet. The approver decides with **Able to Approve Vendors** plus **Able to View Vendors**. Requesters see their intake's overall score and proposed tier, and approvers also see the points behind each answer. The [intake and tiering article](https://support.simplerisk.com/kb/06-06-vendor-intake-and-tiering?hsLang=en) in the User Guide walks through both sides, and you can point your requesters and approvers at it.

One consequence of the tier rule matters on day one: **while the Extra is active, a vendor can't become Active without a tier.** The status control refuses with *A vendor needs a tier before it can become Active. Approve an intake or set a tier first.* Approving an intake moves an Onboarding vendor to Active and sets its tier in one step, and an approver can also set a tier directly from the vendor page. Vendors that were already Active when you activated the Extra keep their status and show as **Untiered**, but one that later goes to Offboarding can return to Active only once it has a tier.

### 8. Set each tier's template and evidence, and the reminders

A tier's policy is part of the tier. In **Vendor tiers** (step 5), the table shows each tier's **Template** and **Required evidence** next to its minimum score and review interval, and the pencil on a row (or **Add tier**) opens one dialog with all five fields: **Tier name**, **Minimum score**, **Review interval**, **Template** and **Required evidence types**. **Save** saves the tier and its policy together. The **Notifications** section, under **Due diligence** in the side navigation, holds the reminder settings.

The **Template** is the questionnaire the vendors of that tier get when their review is due. It's preselected when an analyst selects **Send Assessment** on a vendor, and it's what automatic reassessment sends. Pick a template for the tier, or **None**. A new tier starts with no template.

**Required evidence types** is a set of tick boxes, one per enabled evidence type (step 9 sets the types up). Tick what the vendors of the tier must keep on file. A vendor that has no current item of a ticked type shows it on its **Evidence** tab under **Required for this tier and missing**. A type you later disable stays ticked, labelled *{name} (disabled)*, so a save never drops it without you seeing it, but a disabled type is no longer counted as missing. Contract terms are made required per tier on the terms themselves, in step 10.

Templates come from the Assessments Extra, so the **Template** field leans on the Assessments permission. The required evidence doesn't, so every configurer can change it, and a template you aren't allowed to change is left exactly as it is:

- **Without the Assessments Extra active,** the template field is disabled and the dialog says *Questionnaire templates need the Risk Assessment Extra.* You can still save the required evidence.
- **Without Allow Access to "Assessments" Menu,** the template is read-only and the dialog says *Choosing or changing a questionnaire template needs the Allow Access to "Assessments" Menu permission.* A tier with a template shows *A template is set* instead of the name, because template names are Assessments data. Only a user with the permission can choose, change or clear a template. For anyone else, a save leaves every template as it is, whether it comes from the dialog or the API: whatever template the save names is ignored.
- **A template that was deleted in the Assessments Extra** shows *That questionnaire template no longer exists.* Pick another or **None**. Until you do, sending to that tier's vendors needs a template chosen by hand, and automatic reassessment skips them.

If the tier saves but its template or evidence doesn't (for example because another configurer changed the policies at the same moment), the dialog stays open and says so. The tier is already saved, so selecting **Save** again only retries the policy. The required evidence types have no protection against two configurers saving at once: the last save wins.

Analysts see the same policies, read-only, at the bottom of the **Conduct Due Diligence** page's **Reassessments** view.

#### Notifications

The **Notifications** card has seven settings and a **Save notification settings** button.

#### **Review lead time (days)**

- **Default:** 30
- **What it does:** How many days before its next review date a vendor counts as **Due soon** and its business owner is reminded. A whole number from 0 to 365. The **Conduct Due Diligence** page uses it as its default window.

#### **Send vendor notifications**

- **Default:** On
- **What it does:** The master switch for the daily notification run. It covers the review, evidence, contract and finding reminders alike. While it's off, the run sends nothing at all, automatic reassessment included.

#### **Send reassessment questionnaires automatically**

- **Default:** Off
- **What it does:** Each daily run creates and sends a questionnaire for every vendor that's due or overdue, without anyone selecting **Send Assessment**. Works only while **Send vendor notifications** is on.

#### **Evidence expiry lead time (days)**

- **Default:** 30
- **What it does:** How many days before an evidence item expires it counts as **Expiring soon** and the vendor's business owner is reminded. A whole number from 0 to 365.

#### **Contract renewal lead time (days)**

- **Default:** 60
- **What it does:** How many days before a contract's notice deadline, renewal date or end date its owner is reminded, and the window for the **Renewal due** and **Ending soon** statuses. A whole number from 0 to 365.

#### **Finding due date lead time (days)**

- **Default:** 14
- **What it does:** How many days before its due date an open finding counts as **Due soon** and its owner gets the first reminder. A whole number from 0 to 365. With 0, the owner is reminded on the due date itself.

#### **Maximum file size (MB)**

- **Default:** 25
- **What it does:** The largest evidence or contract file accepted, from 1 to 100 MB. The server's own limits can keep the real limit far lower: on a stock MySQL install it's about 3.5 MB. Step 11 explains why, and the card shows the limit in force beside the field.

Every reminder goes out two ways: as an email and as an in-app notification, which appears in the notification bell with a link to the vendor and expires after 60 days. Both go to the same people, the same number of times, and the one **Send vendor notifications** switch turns both off. A person needs an active account with a usable email address to get either. The daily run emails a vendor's **Business Owner** twice per review date: once when the date comes within the lead time (*Vendor review coming up: {vendor}*) and once when it passes (*Vendor review overdue: {vendor}*). Each email goes once. A record of every email sent is kept, so a second run the same day sends nothing, a run that's retried sends nothing twice, and a mail server failure isn't retried every few minutes. Only Onboarding and Active vendors whose tier has a review interval above 0 are covered. The business owner must be an enabled account with an email address that holds **Allow Access to "Vendor Risk" Menu** and **Able to View Vendors** (or a permission that includes it) and can see the vendor under team separation; otherwise no email goes, and a vendor with no business owner gets none either. The **Conduct Due Diligence** page lists untiered vendors with a review date too, but nobody is emailed about them.

The same run also sends evidence and contract reminders, for vendors that are Onboarding, Active or Offboarding. Offboarding is included on purpose: a vendor you're leaving still holds your data, and its certificates and exit dates still matter.

#### Evidence expiring

- **Sent:** when an item's expiry date comes within the evidence lead time
- **Subject:** *Vendor evidence expiring: {title} ({vendor})*
- **Goes to:** The vendor's business owner

#### Evidence expired

- **Sent:** when the expiry date passes, for up to 30 days afterwards (or the evidence lead time, if that's longer)
- **Subject:** *Vendor evidence expired: {title} ({vendor})*
- **Goes to:** The vendor's business owner

#### Notice deadline coming up

- **Sent:** when the notice deadline comes within the contract lead time
- **Subject:** *Contract notice deadline coming up: {title} ({vendor})*
- **Goes to:** The contract's owner, else the business owner

#### Notice deadline passed

- **Sent:** when the notice deadline passes while the renewal date is still ahead
- **Subject:** *Contract notice deadline passed: {title} ({vendor})*
- **Goes to:** The contract's owner, else the business owner

#### Renewal coming up

- **Sent:** when the renewal date comes within the contract lead time
- **Subject:** *Contract renewal coming up: {title} ({vendor})*
- **Goes to:** The contract's owner, else the business owner

#### Ending soon

- **Sent:** when the end date of a contract with no renewal date comes within the contract lead time
- **Subject:** *Contract ending soon: {title} ({vendor})*
- **Goes to:** The contract's owner, else the business owner

A few rules shape these:

- **The contract's owner, else the vendor's business owner.** A contract with no **Owner** goes to the business owner. A contract whose owner is set but can't receive it goes to nobody: there's no fallback, so a stale owner silences the reminders. Keep owners current when people leave.
- **The recipient must be able to open the item.** For evidence, the business owner needs the menu permission, a view permission and either **Able to Run Vendor Due Diligence** or **Able to Manage Vendor Evidence**. For contracts, the recipient needs the menu permission, a view permission and **Able to View Vendor Contracts** (or **Able to Manage Vendor Contracts**). In both cases they must be an enabled account with an email address that can see the vendor under team separation. A business owner who only holds the review permissions gets review reminders and nothing about evidence or contracts.
- **Renewed evidence goes quiet.** An item stops producing reminders once the vendor holds a newer item of the same type that expires later. Add this year's SOC 2 report and nobody hears that last year's is expiring. An item with no expiry date counts as a renewal only for a type with no default validity, the same rule the **Required for this tier and missing** card uses, so an undated import of a SOC 2 report doesn't silence the dated one.
- **Some dates never produce a reminder.** Evidence with no expiry date, a contract with a notice period of 0 (the renewal reminder covers it), and a contract that has already ended. Renewal and end dates get a "coming up" email only, never a "passed" one.
- **Each email goes once per item, date, stage and recipient,** like the review reminders. Change a renewal or expiry date and a fresh set follows. Change a vendor's business owner or a contract's owner, and the new person gets the overdue emails for items that were already overdue, because they haven't been told yet.

Finding reminders follow the same pattern with two differences: they cover Prospect vendors as well, because a finding from pre-contract due diligence is worth chasing before anyone signs, and their lead time is the finding lead time from step 12, 14 days by default.

#### Finding due soon

- **Sent:** when an open finding's due date comes within the finding lead time
- **Subject:** *Vendor finding due soon: {title} ({vendor})*
- **Goes to:** The finding's owner, else the business owner

#### Finding overdue

- **Sent:** when the due date passes and the finding is still open
- **Subject:** *Vendor finding overdue: {title} ({vendor})*
- **Goes to:** The finding's owner, else the business owner

Only findings that are **Open** or **In Progress** and have a due date are covered, for vendors that are Prospect, Onboarding, Active or Offboarding. A Terminated vendor's findings are never reminded about, even when they're still open. The recipient needs the menu permission, a view permission and either **Able to Run Vendor Due Diligence** or **Able to Manage Vendor Findings**, an enabled account with an email address, and, under team separation, a team that can see the vendor. As with contracts, a finding whose owner can't receive the email goes to nobody: there's no fallback to the business owner. Each email goes once per finding, due date, stage and recipient, so a new due date or a new owner starts the reminders again for that date or person. A finding that is already overdue then gets only the overdue email.

**Old dates don't produce a burst of email.** A review that is already past due is emailed once, and evidence is handled by a window: an expired item is emailed only within 30 days of its expiry date (or the evidence lead time, if longer), so old reports entered for the record, and items that lapsed long before you turned notifications on, send nothing. A notice deadline that passed is emailed only while the renewal date is still ahead. Items that are coming up but not yet overdue do get their "coming up" email on the first run, and an item whose date changes starts a fresh cycle. Reviews that went overdue while **Send vendor notifications** was off are emailed when you turn it back on, and so is evidence that expired within that window.

**Overdue findings have no window at all.** Every open finding past its due date gets one overdue email, however long ago the date passed. Upgrading doesn't cause a burst, because findings arrive with this release and there are none yet. Two things can: turning **Send vendor notifications** back on after a long pause, and people entering a backlog of old findings with due dates already in the past. Each of those findings mails its owner once on the next daily run. If you're about to load a pile of historical findings, tell the owners first, or leave the due dates empty until the findings are in.

**Automatic reassessment sends questionnaires on its own, so turn it on deliberately.** When it's on, the next daily run sends one questionnaire to every eligible vendor that's due soon or overdue, and there's no backlog protection: if forty vendors are overdue on the day you switch it on, forty questionnaires go out. A vendor is eligible when it's Onboarding or Active, its tier has a review interval and a template in its tier policy, and it has at least one contact linked as a SimpleRisk user or an Assessment Contact. Each questionnaire goes to all of those contacts, is created and emailed exactly as a manual send is, and is sent once per review date. A vendor whose earlier questionnaire is still open, or whose questionnaire record shows **Needs attention**, is skipped until that record is resolved; an analyst's manual send isn't blocked by **Needs attention**. A vendor skipped for missing contacts is tried again on the next daily run. Nothing is sent while the Assessments Extra is inactive, and the card says *Automatic reassessment needs the Risk Assessment Extra. Nothing is sent while it is not active.*

Turning automatic reassessment on needs what a manual send needs as well: **Allow Access to "Assessments" Menu**, **Able to Add Questionnaires** and **Able to Send Questionnaires**, because the job then creates and sends questionnaires in the Assessments Extra. A configurer without all three can only turn the switch off; the card says *Turning on automatic reassessment needs the Allow Access to "Assessments" Menu, Able to Add Questionnaires and Able to Send Questionnaires permissions. While it is on, so do turning on Send vendor notifications and raising the review lead time.* That second sentence matters because either change makes the daily run send more questionnaires: while automatic reassessment is on, such a configurer can't turn **Send vendor notifications** back on or raise **Review lead time (days)**, though they can still lower the lead time, turn notifications off, or turn automatic reassessment off in the same save. Once it's on, the daily run sends with SimpleRisk's own authority, not with the permissions of whoever turned it on: it keeps sending until someone turns it off, even if that person later loses those permissions. While **Send vendor notifications** is off, the automatic switch is disabled and keeps its value.

#### How questionnaires and their status flow

Every send, manual or automatic, creates a new questionnaire in the Assessments Extra, named `{vendor} - {template} - {date}`. It belongs to the vendor's teams. A manual send's owner is the person who sent it; an automatic one has no owner, so find it in the Assessments Extra by its name or team. Its own reminder and recurring-send settings are off, so the Assessments Extra never resends it on a schedule that competes with the tier's interval. Tell your Assessments users not to turn those on for these questionnaires. A send goes only to an Onboarding or Active vendor, and only one questionnaire per vendor can be open (**Sent**, **In Progress** or **Awaiting review**) at a time.

The Assessments Extra has no way to tell the Vendor Risk Management Extra a response came in, so SimpleRisk asks it. A background job checks every open questionnaire record every 15 minutes while the Assessments Extra is active, and analysts can refresh one record on demand. A record becomes **Completed** when every response is approved, **Needs attention** when one is rejected or not approved, and **Canceled** when the questionnaire's sends are deleted in the Assessments Extra. Reopening a response there moves the record back to an open status. When a record completes, its review date is the date the last response was approved and the vendor's next review date moves forward from it. The [vendor due diligence article](https://support.simplerisk.com/kb/06-07-vendor-due-diligence?hsLang=en) in the User Guide walks analysts through the vendor side.

### 9. Set up evidence types

The **Evidence types** card lists the kinds of evidence your vendors provide. Each row has a **Name**, a **Default validity**, an **Enabled** switch, an **In use** count, and edit and delete buttons. A fresh activation loads six:

- **SOC 2 Type II report:** 12 months
- **ISO 27001 certificate:** 36 months
- **Penetration test report:** 12 months
- **Cyber insurance certificate:** 12 months
- **Data processing agreement:** Does not expire
- **Business continuity plan:** 12 months

**Add evidence type** asks for a **Name**, unique ignoring case and up to 255 characters, and a **Default validity (months)** from 1 to 120. Leave the validity empty for evidence that doesn't expire, such as a signed agreement. The validity is what fills in an item's expiry date when someone adds evidence with an issue date and no expiry, so a SOC 2 report issued on 15 March expires on the same day a year later. Changing a type's validity affects evidence added afterwards. Items already stored keep their dates.

The **Enabled** switch saves as soon as you flip it. A disabled type keeps its items, isn't offered for new ones, and isn't counted as missing for any tier, though a tier that required it keeps it ticked in **Tier policies**. That makes disabling the way to retire a type. Deleting is refused while anything uses the type, and **In use** shows how many evidence items and tier requirements do. The count covers every vendor, including ones you can't see under team separation, so it can be above 0 when you see no items.

To say which types each tier must have, tick them in the tier's dialog under **Vendor tiers** (step 8). Nothing is required out of the box.

### 10. Set up contract terms

The **Contract terms** card holds the catalog of terms recorded on every contract, in the order shown on the contract form. Each row has **Order** buttons to move it up and down, its **Name** (with a **Standard term** badge for the shipped ones), its **Type**, the tiers it's required for (**Required for tiers**), an **Enabled** switch, and edit and delete buttons.

A term has one of seven types, and the type decides what a preferred position looks like and when a value counts as a deviation:

#### **Free text**

- **A value is:** Up to 5,000 characters
- **Preferred position:** None
- **Deviation when:** Never

#### **Single choice**

- **A value is:** One of 2 to 50 choices
- **Preferred position:** The acceptable choices
- **Deviation when:** The value isn't one of them

#### **Multiple choices**

- **A value is:** One or more of 2 to 50 choices
- **Preferred position:** The acceptable choices
- **Deviation when:** Any chosen value isn't one of them

#### **Yes or no**

- **A value is:** Yes or No
- **Preferred position:** **Yes** or **No**
- **Deviation when:** The answer differs

#### **Number**

- **A value is:** A number, with an optional unit of up to 20 characters
- **Preferred position:** **At least** or **At most** a **Threshold**
- **Deviation when:** The value is on the wrong side of the threshold

#### **Amount and basis**

- **A value is:** An amount and one of 1 to 20 bases, with an optional unit
- **Preferred position:** A **Minimum amount**, with a **Basis** or **Any basis**
- **Deviation when:** The amount is below the minimum, or the basis differs from the one named

#### **Date**

- **A value is:** A date
- **Preferred position:** None
- **Deviation when:** Never

A value exactly at the threshold or the minimum counts as preferred. A term with no preferred position is tracked but never flagged. Two more flags don't depend on the preferred position at all. **Missing** means the term is required for the vendor's tier and the contract has no value. **Not valid** means the stored value no longer fits the term or couldn't be decrypted. The contract's own tab and the terms matrix both show these flags, worked out every time they're read, so a change to a preferred position re-flags every contract straight away.

A fresh activation loads sixteen standard terms, none of them required for any tier:

#### Governing jurisdiction

- **Type:** Single choice: England and Wales; Delaware, USA; New York, USA; California, USA; Ireland; Germany; Other
- **Preferred position:** None

#### Limitation of liability

- **Type:** Amount and basis: Months of fees, Fixed amount
- **Preferred position:** At least 12, Months of fees

#### Indemnification

- **Type:** Single choice: Mutual; Vendor indemnifies us; We indemnify vendor; None
- **Preferred position:** Mutual or Vendor indemnifies us

#### Termination for convenience

- **Type:** Yes or no
- **Preferred position:** Yes

#### Auto-renewal

- **Type:** Yes or no
- **Preferred position:** None

#### Data processing agreement in place

- **Type:** Yes or no
- **Preferred position:** Yes

#### Breach notification window

- **Type:** Number, in hours
- **Preferred position:** At most 72

#### Audit rights

- **Type:** Yes or no
- **Preferred position:** Yes

#### Subprocessor approval

- **Type:** Single choice: Prior approval required; Notice only; None
- **Preferred position:** Prior approval required or Notice only

#### Data return or destruction on exit

- **Type:** Yes or no
- **Preferred position:** Yes

#### Insurance requirements

- **Type:** Free text
- **Preferred position:** None

#### SLA uptime

- **Type:** Number, in percent
- **Preferred position:** At least 99.9

#### Payment terms

- **Type:** Number, in days
- **Preferred position:** None

#### Late payment interest

- **Type:** Number, in percent per month
- **Preferred position:** None

#### Rescheduling fee cap

- **Type:** Number, in percent
- **Preferred position:** None

#### Term length

- **Type:** Number, in years
- **Preferred position:** None

**Limitation of liability has no "Unlimited" basis,** because every amount-and-basis value needs a number. Uncapped liability is recorded by leaving the term empty and saying so in the contract's **Notes**. If that's common in your contracts, tell the people who record them, or they'll invent a convention of their own, such as a cap of 9999 months. Mind one side effect: if a tier requires **Limitation of liability**, an empty value shows as **Missing** on those contracts for good, and they appear whenever someone filters the matrix to flagged terms.

Edit a term with its pencil button. The dialog has the **Name** (unique ignoring case, up to 255 characters), a **Description** of up to 1,000 characters, the **Type**, the choices or bases with a **Unit** where the type has one, the **Preferred position**, and **Required for tiers**: the tiers whose vendors' contracts show the term as **Missing** until it has a value. **Add contract term** opens the same dialog empty.

The catalog protects values that contracts already hold:

- **The type can't change while any contract holds a value for the term.** The type select is disabled and says *The type cannot change while {count} contract(s) hold a value for this term. To use another type, disable this term and add a new one.*
- **A choice or basis that values use can't be removed or renamed.** The save is refused with *This term holds values on {count} contract(s), so it cannot be deleted, change type, or lose a choice or basis those values use. Disable it instead.* Adding choices is always fine.
- **The count is contract data.** How many contracts use a choice is worked out from their values, so only a configurer who also holds **Able to View Vendor Contracts** sees the number. Anyone else gets the same refusal without it: *Contracts hold values for this term, so it cannot be deleted, change type, or lose a choice or basis those values use. Disable it instead.* The type select then says *The type cannot change while contracts hold a value for this term. To use another type, disable this term and add a new one.*
- **A term with values can't be deleted,** and **a standard term can't be deleted at all.** Both can be disabled. A disabled term keeps its values but disappears from every contract's terms and from the matrix.

Checking whether a choice is in use means decrypting every value of that term, so on an instance with thousands of contracts, that save takes noticeably longer than the others.

### 11. Understand the file size limit

Evidence files and contract documents have their own size limit, separate from the **Maximum Upload File Size** that the rest of SimpleRisk uses. It's the smallest of four numbers:

1. **Maximum file size (MB)** on the **Notifications** card: 25 by default, from 1 to 100.
2. PHP's `upload_max_filesize` and `post_max_size`.
3. An eighth of the memory PHP has left under `memory_limit` while it handles the upload. Encrypting a file and reading it back takes several copies of it in memory, and the divisor keeps a large file from exhausting PHP part-way through.
4. About three-quarters of the largest file the database can store, less a small margin. Once encryption is in effect (see "Vendor details and encryption"), the encrypted form of a file is about a third larger than the file.

The card shows the result beside the field: *Right now, files up to {size} MB can be uploaded. The server's upload and database limits can keep this below the maximum.* People who can upload evidence or contract documents see the same figure next to their file field. On a stock MySQL install, the default `innodb_log_file_size` of 48M lets the database store files of about 4.8 MB, so evidence tops out at about 3.5 MB whatever you set here. To accept 25 MB files, the database has to store about 34 MB, which means an `innodb_log_file_size` of about 340M, ten times the largest file, and PHP needs about 200 MB of memory free above what the request already uses. Treat a `memory_limit` of 256M as a floor, not a target, and check the figure on the card after each change. [File Upload Size Limits](https://support.simplerisk.com/kb/12-07-file-upload-size-limits?hsLang=en) explains how to raise the PHP and MySQL values.

File types follow SimpleRisk's own **Allowed File Types** list under **File Upload**. SimpleRisk checks the file's actual content, not just its extension, so a renamed file is refused with *This file type is not allowed. Upload a file of a type listed under Allowed File Types.* A file over the limit is refused with *The uploaded file was too big.* In both cases nothing is stored. The same checks apply to a file someone adds from a questionnaire answer. Once added, that file is ordinary evidence: everyone who can read the vendor's evidence can download it, including holders of **Able to Run Vendor Due Diligence** who have no access to the Assessments Extra.

### 12. Check the scales findings are scored on

Analysts record findings on a vendor's **Findings** tab, and the [findings and offboarding article](https://support.simplerisk.com/kb/06-09-vendor-findings-and-offboarding?hsLang=en) in the User Guide walks them through it. There is no finding setting card. A finding is scored the way a risk is: the analyst picks a **Likelihood** and an **Impact**, and SimpleRisk works out the score and the risk level from your Classic scoring on the **Risk Configuration** page. The level names and colors are yours, and a vendor's tier doesn't change the score.

Two things follow for you. First, change the Classic likelihood and impact scales or the risk levels and findings are re-read against them at once, so no regrading step exists. Second, a finding whose stored likelihood or impact is no longer on the scale can't be edited or sent to the risk register until someone picks levels that exist. Check the findings list after you change the scales.

The one finding-related timing setting, **Finding due date lead time (days)**, lives on the **Notifications** card (step 8). When someone sends a finding to the risk register with **Create risk**, the dialog starts with the finding's own likelihood and impact.

### 13. Set up the offboarding checklist

The **Offboarding checklist** card holds the steps every vendor works through between Offboarding and Terminated. A fresh activation loads three, all required and enabled:

- **Access revoked:** Every account, key and network path the vendor had into our systems has been removed.
- **Data returned or destroyed:** The vendor has returned or destroyed our data and confirmed it did.
- **Final attestation received:** The vendor has confirmed in writing that the offboarding is complete.

Each row has **Order** buttons to move it up and down, its **Name** and description, **Required** (Yes or No), an **Enabled** switch, and edit and delete buttons. **Add checklist item** opens a dialog with the **Name** (up to 255 characters, unique ignoring case), a **Description** of up to 1,000 characters and a **Required** box: *A vendor cannot be terminated while a required step is not done.* The list holds at most 50 items. The **Enabled** switch saves as soon as you flip it, and a disabled item isn't added to new checklists.

A vendor gets its checklist when it moves from Active to Offboarding. SimpleRisk copies every enabled item into it at that moment, with its name and whether it's required, so later changes to the catalog never touch a checklist that has started. That's why deleting an item is always allowed: *Delete the checklist item "{name}"? Checklists already started keep their copy of it. To stop offering it without deleting it, disable it instead.* A vendor that goes back to Active and is offboarded again later gets a new checklist with every step undone. With no enabled items, a vendor's checklist is empty and counts as done, so it can be terminated as soon as someone who can edit and approve vendors does so; the catalog is what makes offboarding a real gate.

Terminating a vendor that's in Offboarding then needs three things, wherever it's done (the vendor's **Offboarding** tab, the **Change Status** card or the API): **Able to Edit Vendors**, **Able to Approve Vendors**, and a started checklist with every required step done. A checklist with no required steps counts as done once it's started. Terminating a Prospect or Onboarding vendor works as it always did. Open findings don't block termination; once the vendor is Terminated they stay listed and stop producing reminders.

**Vendors already in Offboarding have no checklist.** That's every vendor that was in Offboarding when you upgraded to this release or activated the Extra, including a reactivation. They can't be terminated until someone with **Able to Manage Vendors** opens the vendor's **Offboarding** tab and selects **Start checklist**. That button appears only for an Offboarding vendor with no checklist, and it never resets one that exists. Check **Manage Vendors** filtered to Offboarding after the upgrade and start them, or your approvers will meet *This vendor cannot be terminated until its offboarding checklist is started and every required step is done.* with nobody sure why.

While the Extra is deactivated, none of this applies: SimpleRisk's own rules let **Able to Edit Vendors** alone terminate an Offboarding vendor, with no checklist and no approver, just as they let it make a vendor Active.

### 14. Choose the controls each tier requires

Control coverage asks, vendor by vendor, whether a vendor meets the controls you hold its tier to. You choose the controls in the tier's own dialog: under **Vendor tiers**, select the pencil on a tier (or **Add tier**) and use the **Required controls** field. The field appears for a user who holds **Able to Configure Vendor Risk Management** and **Allow Access to "Governance" Menu**; without both, the **Required controls** column and field aren't shown.

Any control in SimpleRisk can be required, from any framework, and custom controls you created yourself too. The Secure Controls Framework (SCF) Extra isn't needed. The table shows how many controls each tier requires, and a click on the row lists them.

**Choose controls** opens the same picker SimpleRisk uses for documents, risks and tests: a Framework column and a Control family column that narrow the list, a search box (by number or name), the controls, and a Selected column. A control that sits in several frameworks appears once. Tick the controls the tier requires, select **Use these controls**, and the dialog shows them as chips; a chip's cross takes one off. **Save** saves the tier, its template and evidence, and its required controls together. If the tier saves but the controls don't, the dialog stays open and says so, and another **Save** retries just the controls.

A tier can require at most 200 controls. The usual set is much smaller: start with the handful that match how you actually assess vendors, such as third-party risk assessments, contract requirements, and the review of third-party services, rather than a whole family. Every required control is a row an analyst has to record for every vendor in the tier, and a list of forty controls that nobody records produces a coverage report that's all **Not assessed**. A save is refused if it names a control that doesn't exist or has been deleted.

If a control a tier requires is later deleted, the dialog no longer lists it and the next save of that tier drops it. Removed controls count nowhere, so nothing is lost by removing them.

Requiring controls changes nothing on its own. Each vendor in the tier gains the controls on its **Controls** tab as **Not assessed**, analysts record them there, and the [overview and control coverage article](https://support.simplerisk.com/kb/06-10-vendor-overview-reports-and-control-coverage?hsLang=en) in the User Guide walks them through it. Coverage is what the analyst records; it isn't worked out from questionnaire answers. A waiver links a control to one of SimpleRisk's own exceptions, and an approved exception in the open state waives its control for every vendor that links it, including after its review date. Tell the people who approve exceptions, because their approval now reaches vendors too.

### 15. Understand the Vendor Risk Dashboard and Reports

The **Vendor Risk Dashboard** and the six vendor reports need no setup. The reports live on the Reports page, and each is its own page: **Vendor Register** and **Fourth-Party Concentration** for anyone who can view the Extra's data, **Overdue Vendor Reassessments** and **Vendor Risk Heat Map** for users who can run due diligence, **Expiring Vendor Evidence and Contracts** for users who can read evidence or contracts, and **Vendor Control Coverage** for users who can run due diligence and also hold **Allow Access to "Governance" Menu**. A report's card is there only while the Extra is active and only for those users; if the Extra is deactivated, the cards and the pages go with it, and an old address sends the user back to the Reports page. The old **Vendor Reports** address still works and opens the vendor register. The dashboard is made of widgets (vendor counts, reviews overdue, evidence expiring, contracts due, open findings, control gaps, vendors by tier and status, and vendors that need attention). With the Customization Extra active, a user can rearrange, resize and add widgets with **Edit Layout** and an administrator can set the default for everyone; with the Import/Export Extra active, **Export PDF** saves it as a PDF. A widget whose figures the user may not read says so instead of showing zero. They work from the data the rest of the Extra already holds and the lead times on the **Notifications** card: the evidence and contract lead times set what counts as expiring and the default windows on the expiring evidence and contracts report. Neither the dashboard nor the reports add a setting or a reminder.

Every figure on the dashboard and in the reports, and every row of the register and its export, counts only the vendors the user can see, worked out fresh for each request. There's no shared cache, because two users rarely see the same vendors.

The reports have a ceiling. A request that would cover more than 20,000 vendors the user can see, or that would use more than 60% of PHP's `memory_limit`, is refused as a whole: *This report would include more vendors than one report can cover. Where the report has filters, narrow it by status, tier or name.* or *This report needs more memory than the server allows. Filter it to fewer vendors, or ask an administrator to raise PHP's memory limit.* The Overview has no filters, so above the ceiling it's refused entirely rather than shown in part, while the register and its export can be narrowed. Our estimate is about 60 MB of memory for one Overview at 20,000 vendors. The export writes in batches of 500 vendors and never holds the whole register at once. If your installation is anywhere near that size, give PHP room and check the pages as an administrator, who sees every vendor.

The export is a CSV or XLSX file named `vendor_register_{date}`, with only the column groups the user can read. It never contains contract values, services, notes or any other encrypted field, so it never needs the encryption key once encryption is available. A CSV cell that would start with `=`, `+`, `-`, `@`, a tab or a carriage return gets a leading apostrophe, so no spreadsheet runs it as a formula; XLSX cells are stored as text. Headers and the status, review state, outcome and tier source cells are in the exporting user's language.

The dashboard also offers a **Vendors by risk level** chart, a pie of the vendors by their computed risk level with a slice for the unscored ones. Like every widget on the dashboard it is offered, in the picker and in a default layout, only to a user who may read what it shows, and this one needs the computed score (see step 18). A user who has already saved a layout keeps it, so a widget that was added later doesn't appear for them until they restore the default layout or add it from the picker.

### 16. Know the rules for fourth-party dependencies

A dependency records that one vendor relies on another for a service. The fourth party must already be a vendor, so adding a supplier your vendor relies on means adding it on **Manage Vendors** first, where it starts as a Prospect and counts as a vendor everywhere. That's deliberate: you can then tier it, review it and record findings against it like any other supplier. Dependencies need no setup and have no settings. Reading them needs **Able to View Vendors**; adding, editing and deleting them needs **Able to Manage Vendors**, and both vendors must be ones the user can see.

The rules analysts meet:

- A vendor can't depend on itself, and the same vendor, fourth party and service can be recorded only once. The service comparison ignores letter case and accents, so "Café" and "cafe" are the same service. It also treats any two emoji as equal, which occasionally refuses a harmless second row.
- A vendor can have at most 200 dependencies. The limit counts dependencies on vendors the user can't see, so under team separation a user can be refused while fewer show, and can't remove the hidden ones themselves.
- Cycles are allowed, because real supply chains have them. Nothing follows a chain past one step, so a cycle can't loop.
- **Service involved** is plain text of up to 255 characters, and it appears in the concentration report. **Notes** hold up to 5,000 characters and are protected like the other sensitive fields (see "Vendor details and encryption").

### 17. Set up contract term extraction

Term extraction has an AI provider read a contract's PDF or Word (`.docx`) document and propose values for the enabled contract terms and for the contract's start, end and renewal dates, notice period, total value, currency and SLA summary. Analysts check each proposal against the passage it quotes and accept, edit or reject it; the [term extraction article](https://support.simplerisk.com/kb/06-11-vendor-contract-term-extraction?hsLang=en) in the User Guide walks them through it. Your part is the provider, the permissions (step 2), and the **Contract term extraction** card at the bottom of the Extra's page.

**The provider.** Extraction uses the AI provider configured in SimpleRisk itself: the provider, model and API key on the **Provider Configuration** tab, which [Configuring AI Providers](https://support.simplerisk.com/kb/10-05-configuring-ai-providers?hsLang=en) explains. The AI Extra isn't needed. Extraction is switched on by an administrator under **Artificial Intelligence**, then **Capabilities**, where it appears as **Vendor Contract Term Extraction** (off until you turn it on). Until it is on, the **Extract terms** button and the **Contract term extraction** settings stay hidden. Until a provider is configured, nobody sees **Extract terms**, and the card and the proposals panel show *The AI provider is not configured. Unable to produce live results until you set the provider and API key on the Provider Configuration tab.* A model with a large context window works best, because a long contract is sent in parts of about 48,000 characters.

**The card.** It holds three settings and **Save extraction settings**:

#### **Model calls allowed per contract document**

- **Default:** 20
- **Range:** 1 to 200
- **What it limits:** Calls to the provider, summed over every extraction of the same document

#### **Tokens allowed per contract document**

- **Default:** 200,000
- **Range:** 10,000 to 5,000,000
- **What it limits:** Tokens sent and received, summed the same way

#### **Extractions each user may request per hour**

- **Default:** 10
- **Range:** 1 to 100
- **What it limits:** Requests per user in any rolling hour

The two budgets belong to the document, not the contract: replacing a contract's document starts a fresh budget, and running the same document again draws on what's left. A short contract with the sixteen standard terms is one call. A long one costs one call per part of the text for each group of up to 20 terms, and a reply the Extra can't read gets one retry, which counts as a call. Texts longer than about 384,000 characters (roughly 100 pages) are refused outright.

The token budget is checked twice. Before anything is sent, SimpleRisk plans the whole run with an estimate of four characters per token and refuses the run if the plan doesn't fit in what's left. Then, before every call, it checks the actual count so far. The estimate suits English and other Latin-script text. Chinese, Japanese, Arabic, Cyrillic and similar scripts use more tokens per character, so runs on those contracts can stop with the budget message part-way, having already spent the calls they made. Budget accordingly if your contracts are in those languages.

The hourly limit counts every request, including one that fails, is cancelled or is interrupted, and one that couldn't be queued. Deleting the contract or the vendor doesn't give requests back.

Under the settings the card shows a usage line: *Usage in the last 30 days, across all users and vendors. Extractions: {runs}. Model calls: {calls}. Tokens: {tokens}.* It covers the whole instance, including vendors the reader can't see under team separation; it shows counts only, never a vendor or a contract. Use it to check the bill your provider sends, with two caveats:

- Some providers don't report token usage, and the OpenAI-compatible ones are among them. Their calls are counted at the same four-characters-per-token estimate, so the line can be well off for non-Latin text.
- One counted call can hide up to three requests to the provider, because a rate-limit or overload reply is retried within the call. A retried request the provider still bills isn't counted again, so usage can read low by up to three times in a bad hour. A call that times out isn't retried and is charged only its input estimate, though the provider may bill the reply it was still writing. Calls that fail before anything is sent are charged an estimate, so the line can also read slightly high.

**What leaves your instance.** The card states it, and the confirmation each user sees before a run says the same:

- **Sent to the provider:** the text of the contract's document, and an internal number and the name, description (up to 300 characters), type, choices and unit of each enabled contract term, plus the names of the four date fields and the three contract fields (total value, currency and SLA summary) and SimpleRisk's fixed instructions. The document's text is the contract itself, so it normally names both parties, their signatories and the amounts; anyone deciding whether a contract may go to a third-party provider should judge it on that basis. Nothing else from SimpleRisk's records goes: not the vendor record's name or ID, the contract's title, its stored value, currency, SLA summary or notes (the AI proposes those from the document and is never shown what you already entered), or anything about the user. Review your term descriptions with that in mind.
- **Stored (encrypted once built-in encryption is available):** each proposed value and the short passage quoted with it, cut to 1,000 characters. The record of a run holds IDs, counts, status and the reason a run failed, never contract text. The extracted text and the AI's full reply aren't kept anywhere.
- **Logged: nothing from the contract.** The extraction logs run numbers, counts, outcomes, status codes and the kind of error only. Contract text, quotes, values, prompts and provider replies are never written to the debug log, at any log level, and neither are provider error messages. SimpleRisk's AI connection runs in a quiet mode for extraction so that its own debug logging leaves them out too.
- **On disk, briefly:** while the document is read, a decrypted copy exists as a temporary file in the server's shared system temporary folder, readable only by the account SimpleRisk runs as. It's deleted when reading ends, also after an error or a PHP fatal error. It is not deleted if the worker process is killed outright: by `SIGKILL`, by a `SIGTERM` it doesn't handle, by the operating system's out-of-memory killer, or by a supervisor stopping or restarting it during a deploy. The copy then stays until the operating system clears its temporary folder. If that matters to you, make sure the temporary folder is cleared on reboot, and avoid restarting workers while an extraction is running.

Every request, cancellation, finished run and decision on a proposal is recorded in the audit trail (see below).

**Size limits.** Three numbers are in play, and the smallest wins:

- the Extra's upload limit (step 11), 25 MB by default;
- the document reader's own limit per type, 24 MB for a PDF and 16 MB for a Word file, which is what the refusal message quotes;
- the memory the background worker has free. Reading a document takes up to about 32 times its size in memory, so a worker with a 512 MB `memory_limit` reads files of up to about 15 MB, whichever type.

So with the defaults, a 20 MB PDF uploads fine and is then refused for extraction with *This document is too large to extract terms from.* Raise the worker's `memory_limit` if your contracts are larger, or lower the upload limit to match what can be read. The request itself checks only the file's size against the reader's limit, so a file over the worker's memory is accepted and its run then fails with the same message.

Word files need one more check, because a `.docx` is a compressed archive. A small file can unpack into an enormous document, and the part of memory used to parse it sits outside PHP's `memory_limit`, so the limits above can't contain it. Before a Word file is parsed, the worker reads how large its text part says it will unpack to, and refuses the run as too large when that's over 16 MB (or over an eighth of the worker's free memory). A file whose text part can't be found at all is refused as not a readable Word document, and so is an archive with thousands of entries or unusually long entry headers, since a real `.docx` has a few hundred at most. A typed contract within the 384,000-character ceiling normally needs far less, but a file converted from PDF or carrying heavy tracked changes can average more than about 40 bytes of markup per character and be refused although its text would fit; resave it as a clean `.docx` or upload the PDF. Word files still use memory outside PHP's limit while they're parsed. In the worst case, a crafted file that sits just under the 16 MB limit could need roughly 0.6 GB beyond `memory_limit` (an estimate, not a measurement), so size the container or server with that headroom.

**The background worker.** Extraction runs as a queued job on the queue worker, ahead of default-priority work. It's slow by design: each call to the provider is allowed 120 seconds per attempt and up to three attempts, so a run that uses its full budget of 20 calls can hold one worker for more than two hours in the worst case. On an instance with a single worker, email and other queued jobs wait behind it. Run more than one worker if extraction is busy. Reading the document has no time limit of its own: a malformed or pathological PDF can keep the parser busy, and the worker with it, for as long as parsing takes. The run's lease (below) then lapses and the run ends as interrupted without sending anything, but the worker is only freed when parsing ends.

A run holds a ten-minute lease that each call renews. Before every call the worker also checks again that the person who requested the run still holds the extraction permissions and can see the vendor, and that the provider is still configured; if either has gone, the run stops there and nothing more is sent. If the worker dies or restarts mid-run, the lease runs out and the run is marked interrupted the next time anyone looks at it. It's never resumed, because that would send the document again unasked; the user requests it again, which costs one request against their hourly limit, and the calls already made stay on the document's budget. A document so large that reading it takes more than ten minutes can end the same way. A run that sits waiting for the worker, because the worker isn't running or its queue entry was lost, never times out: the user can cancel it with **Cancel extraction**, and nothing is sent.

### 18. Set up the risk score weights

Every vendor with a tier has a computed risk score on SimpleRisk's own scale: the tier decides the impact, and the likelihood comes from five factors that measure the vendor against what you require of it. The score is worked out for you; the one thing you decide is how much each factor counts.

Open the Extra's settings page, choose **Risk scoring** in the **Requirements** group (it comes after **Vendor tiers**), and you'll find **Risk score weights** with one whole number from 0 to 100 for each factor:

- **Due diligence:** 25
- **Required controls:** 25
- **Required evidence:** 15
- **Open findings:** 25
- **Contracts:** 10

The weights are relative, not percentages that must add up to 100. A factor that doesn't apply to a vendor (for example controls, when the vendor's tier requires none) is left out and the others are scaled to fill its place. Under the fields, **Share of the score when every factor applies** shows what each factor would count for with the numbers you've typed, so you can see the effect before you save. A weight of 0 takes a factor out of the score altogether. At least one weight must be above zero; saving all zeros is refused with *At least one score weight must be above zero.* and a value outside 0 to 100 with *A score weight must be a whole number from 0 to 100.* beside the field.

Select **Save risk score weights**. If any weight changed, SimpleRisk recalculates the score of every vendor that isn't Terminated before the save returns, so on an instance with many vendors the button can take a moment. A vendor whose score moves because of the new weights gets a history entry labeled **Score settings** saying the weights were changed, so nobody has to guess why a score moved. A vendor whose score stays the same gets no entry. A save that changes nothing recalculates nothing. Terminated vendors keep the last score they had.

Scores also change on their own, as the facts behind them change: a review, a piece of evidence, a contract, a finding, control coverage, a tier or a status. A queued job re-checks every vendor once a day as well, so changes that depend only on the calendar (a review that has become overdue, evidence that is about to expire, a contract that is about to end) appear without anyone editing the vendor. That job runs on the queue worker, so the queue cron jobs from [Before you start](https://support.simplerisk.com/kb/12-09-the-vendor-risk-management-extra#before-you-start) must be running, and it is one more of the Extra's jobs to restart the long-running workers for after an upgrade.

Only a user who can read everything the score is built from sees it: **Able to View Vendors**, **Able to Run Vendor Due Diligence**, **Able to View Vendor Contracts**, the evidence and findings reads those imply, and **Allow Access to "Governance" Menu**, because control coverage is part of the score. The score does not need the control catalog (the Secure Controls Framework (SCF) Extra): when the catalog is not available, the controls factor is left out and the other factors share the weight. Without all of them there's no **Risk** tab on a vendor, no **Risk Score** or **Risk Level** column in the list or the register, no score in the export, no **Vendors by risk level** chart, and the score route answers 403. Granting a user the weights card (**Able to Configure Vendor Risk Management**) doesn't show them any vendor's score.

## Control coverage when the SCF Extra changes

Required controls and control assessments point at SimpleRisk's own copy of each control. For controls that came from the SCF Extra, what happens to them when the SCF Extra changes depends on the change (controls from other frameworks, and ones you created, aren't affected):

- **Installing or updating** makes coverage unavailable for the length of the run, with the busy note above. The rows come back unchanged afterwards.
- **A renamed SCF control** keeps its identity, so requirements and assessments carry on as before.
- **Merged SCF controls** carry their requirements and assessments to the surviving control. When a vendor already has an assessment of the surviving control, the assessment of the merged-away control isn't overwritten: SimpleRisk keeps the merged-away control in retired form, and that assessment shows as a removed control on the vendor's **Controls** tab until an analyst clears it. A waiver doesn't follow a merge: the exception still belongs to the merged-away control, so the assessment on the surviving control no longer counts as waived. Create or edit an exception for the surviving control and link it again.
- **Uninstalling the SCF Extra deletes its controls.** Reinstalling it creates new controls, so every requirement and every assessment recorded before the uninstall becomes **Control removed**, permanently. Nothing remaps them by SCF number. Admins remove the removed requirements on the **Required controls by tier** card, and analysts clear the removed assessments with **Clear** on each vendor's **Controls** tab. While the SCF Extra is uninstalled, every coverage action, **Clear** included, is unavailable, so the clean-up can only happen after you reinstall it. If you never do, nothing can clear those rows from the screens, but nothing counts them either, and they go when the vendor is deleted or the Extra is deactivated. Plan an SCF uninstall like a schema change: tell your analysts first, and expect to re-record coverage afterwards.

One rare case to know about. On MySQL 5.7, a deleted control's ID can be reused by a control created after a database restart, if it was among the most recently created controls (MySQL 5.7 restarts numbering just above the highest ID still in use), and an orphaned requirement or assessment would then point at that unrelated control. A deleted exception's ID can be reused the same way, and an approved, open exception for the same control would then waive it. MySQL 8 doesn't reuse IDs this way. If you're on 5.7 and have just uninstalled and reinstalled the SCF Extra, clear the removed rows promptly.

## Team visibility

Team visibility is SimpleRisk's own vendor rule, and the Extra follows it everywhere: tiers, intakes, contacts, risk scores, assessment records, evidence, contracts, findings, control assessments and offboarding checklists belong to a vendor and are visible exactly when the vendor is. A contract or a finding has no teams of its own. The **Conduct Due Diligence**, **Manage Contracts** and **Manage Findings** pages list only vendors, records, evidence, contracts and findings the user can see, and the terms matrix leaves out every contract of a hidden vendor. The **Vendor Risk Dashboard** and the vendor reports count only visible vendors in every figure, chart, report, register row and export row. A dependency is visible only when both of its vendors are: one whose other end is hidden isn't listed or counted anywhere, so a hidden vendor never shows up as someone's fourth party. A vendor's teams only restrict who sees it when the Team-Based Separation Extra is active. Then:

- Administrators see every vendor.
- A vendor with no teams is visible to every user who can open Manage Vendors. There is no setting to change that for vendors.
- Any other vendor is visible only to users who share at least one of its teams.
- When a non-administrator creates a vendor, they must pick at least one team, and they can only pick teams they belong to.
- When a non-administrator edits a vendor's teams, they manage only their own part of the set. Teams on the vendor that they don't belong to stay as they were.
- A vendor the user can't see behaves exactly like one that doesn't exist: the page says it doesn't exist or isn't accessible, and the same applies to its tier, intakes, contacts, risk score, assessment records, evidence, contracts and their files, findings, dependencies, control assessments and offboarding steps. The one exception is the limit of 200 dependencies per vendor, which counts dependencies on hidden vendors too.
- A user sees only the linked assets and risks they could see anyway. Links to ones they can't see are kept when they save.

A risk created from a finding is a normal risk with its own teams: it gets the vendor's teams at the moment it's created, and changing the vendor's teams later doesn't change the risk's. Who can see the risk follows SimpleRisk's risk rules from then on, so a findings manager can create a risk they can't open themselves. The finding then says so and names the risk's ID, so a risk manager who can see it can pick it up.

Removing a team from SimpleRisk doesn't take it off vendors. That's deliberate: if it did, a vendor whose only team was deleted would become teamless and so visible to everyone.

## Vendor details and encryption

**Availability note.** In this release the Extra stores these fields as entered. Vendor Risk does not use the Encrypted Database Extra, and activating that Extra does not change this. It will use SimpleRisk's own built-in encryption once that ships, and the protection described in this section applies from then on. Until then, treat the contents of these fields as unencrypted in the database and its backups, and don't enter secrets in them. Values saved before the switch stay readable afterwards.

These fields are the ones that will be encrypted before they're stored once built-in encryption is available: a vendor's **Services** text, an external contact's name, e-mail address and phone number, the **Notes** a requester adds to an intake, the **Notes** on an assessment record, and the free-text reasons: the one an approver gives when rejecting an intake or choosing a different tier, and the one recorded when someone sets a tier directly. So are every evidence file and contract document with its file name, an evidence item's **Notes**, a contract's **Contract value**, **SLA summary** and **Notes**, every contract term value, a finding's **Description** and **Remediation notes**, the notes on an offboarding step, a dependency's **Notes**, a control assessment's **Notes**, and each AI term proposal's value and quoted passage. The protection will use SimpleRisk's built-in encryption key, which lives in the key directory that the [encryption articles](https://support.simplerisk.com/kb/09-03-key-management-and-rotation?hsLang=en) tell you to back up. It never depends on the Encrypted Database Extra.

Once built-in encryption is available, if the key is missing or unusable, the Extra still activates and the register still works, but saving one of those fields fails. The user sees *This information could not be saved because the encryption key is unavailable. Ask your administrator to restore the encryption key.* Rejecting an intake, choosing a tier other than the proposed one and setting a tier directly all need a reason, so they are refused without the key. Approving the proposed tier with no reason does not need it. Adding or replacing a file fails the same way, and the whole save is refused: no evidence item, contract or file is left half-stored. Nothing is stored in plain text as a fallback, and over the API the same failure is a 412 response. A download needs the key too. Without it, the download is refused with *This content could not be read because the encryption key is unavailable.* and no file is sent, never the encrypted bytes. Values already stored that can't be read show *This content could not be read because the encryption key is unavailable.* in place of the text. Editing a vendor does not overwrite them.

The vendor record itself (name, legal name, website and the rest of the profile) is not encrypted, which is what lets **Manage Vendors** search by name and legal name. Evidence, contract and finding titles, their dates, an evidence item's type, a contract's notice period, currency and owner, and a finding's likelihood, impact, status, source and owner aren't encrypted either. Neither is a dependency's **Service involved**, nor a control assessment's status and date. That's what lets the expiry calendar, the contract list, the findings lists and the reports work without decrypting anything, and it means titles appear in reminder email subjects. A finding's title also becomes the subject of a risk created from it. Keep secrets out of titles. None of the encrypted fields are searchable, and the terms matrix decrypts term values as it reads them, for the contracts the user can see.

## Vendor changes in the audit trail

Vendor changes go to SimpleRisk's audit trail, which administrators read on the **Audit Trail** page. SimpleRisk itself records creating, editing, renaming and deleting a vendor, every status change, changes to a vendor's linked assets and risks, and the changes it makes to vendors when a category or data type is deleted. The Extra adds every tier change, submitting, approving and rejecting an intake, adding, editing and removing a contact, adding, editing and deleting an assessment record, sending a due diligence questionnaire, whether an analyst sent it or automatic reassessment did, adding (including from a questionnaire answer), editing and deleting evidence, adding, editing and deleting a contract and saving its term values, adding, editing, deleting and changing the status of a finding, creating a risk from a finding and linking or unlinking one, starting an offboarding checklist, ticking or unticking a step, adding, editing and removing a fourth-party dependency, saving or clearing a control assessment and linking or unlinking its exception, and requesting, cancelling and finishing a term extraction and accepting or rejecting each of its proposals. The finished entry is recorded under the person who requested the run, and accepting a proposal whose value the contract already held is still recorded as an accept, together with the usual "terms were updated" entry (or "contract was updated", for a date), and the term records the accepting user as its last editor. Read that pair as a confirmation, not a change of value. Saving tier policies, notification, finding, extraction or risk score weight settings, a recalculated risk score, changes to the evidence types, contract terms, offboarding checklist items and the controls a tier requires, and the status changes the questionnaire check makes, aren't recorded. Each entry records who made the change and when.

The entries name the vendor and the plain values involved: the old and new status, the old and new tier, the contact type and the plain values of the change. They never contain encrypted content or reasons. Assessment record, evidence, contract, finding and offboarding entries name only the record and the vendor by number, never the score, outcome, notes, value, terms, a finding's title or description, or a step's name or notes. A finding's entries add its likelihood and impact and, for a status change, the old and new status; risk entries add the risk's ID. Dependency entries name the dependency and the other vendor by number, never the service or notes, and are recorded on the vendor that depends. Extraction entries name the run or proposal, the contract and the vendor by number, plus the term or date field a proposal was for, never a value or a quote. Control entries name the vendor, the control and any exception by number, plus the status code for an assessment, never the notes or the exception's name, so an audit reader without **Able to View Exceptions** sees exception IDs but not what they are. A contact's name, e-mail address and phone number, an intake's notes, an assessment record's notes and the reasons an approver writes stay out of the audit trail, and you read them on the vendor's own page.

## Vendor names are unique

A vendor's name must be unique across the whole instance, ignoring case. Saving a second vendor with a name already in use returns *The specified name is already in use.*

That check looks at every vendor, including those in teams the user can't see. A user who gets the message for a vendor they can't find in **Manage Vendors** is seeing the rule work as designed, and the answer is to pick a different name or ask an administrator.

## What deactivation deletes

Deactivating the Extra permanently deletes the data it adds, and there is no undo. The page warns you in red, and the **Deactivate** button asks for confirmation. Deactivating through the API has no confirmation prompt: only the admin page warns, so a script that calls the deactivate route deletes the data immediately.

Gone after deactivation:

- every vendor's services description, next review date, tier and tier history;
- the tiers, the intake questions and their answer options, and every intake with its answers, including any edits you made to the defaults;
- every vendor's contacts, computed risk score and score history;
- every evidence file and evidence item, every contract with its document and term values, and the evidence type and contract term catalogs, including any edits you made to the shipped ones;
- every finding, every offboarding checklist with its steps, and the offboarding item list, including any edits you made to the shipped items;
- every fourth-party dependency, the controls each tier requires, and every control assessment with its link to an exception;
- every term extraction run with its proposals, the record of extraction usage behind the hourly limit and the usage line, and the three extraction settings;
- the tier policies and their required evidence types, every assessment record, the record of which review, evidence and contract reminders were sent, and the **Notifications** settings, including the lead times and the maximum file size, and the finding settings;
- the **Vendor Risk Management** permissions, which also removes them from every role and user that held them.

What stays is SimpleRisk's: the vendors themselves, their status, categories, data types, teams and tags, their links to assets, asset groups and risks, the category and data type lists, and the Vendor Management permissions. Risks created from findings stay in the risk register, still linked to their vendor; only the finding behind them is gone. The same goes for exceptions linked as waivers: they stay in Governance, and only the link to the vendor is gone. Vendors that were added only to record them as fourth parties stay too, as ordinary vendors.

Deleting a single vendor on **Manage Vendors** removes the same kinds of Extra data for that vendor alone, including its dependencies in both directions: the ones it had on other vendors, and the ones other vendors had on it. Its control assessments go with it; the controls its tier requires belong to the tier and stay. Its contracts' extraction runs and proposals go too, as they do when a single contract is deleted, but the record of extraction usage doesn't: it stays, so that deleting a contract and extracting again never resets a user's hourly limit or the 30-day usage line. Deleting a contract term deletes its pending and decided proposals with it. Your assessment contacts, assets and users aren't touched, and neither are the questionnaires due diligence created in the Assessments Extra: they stay there, with their responses, but nothing links them to a vendor any more.

Two consequences are worth planning for. The rules for making a vendor Active or Terminated fall back to SimpleRisk's: **Able to Edit Vendors** alone moves a vendor to Active, with no tier and no approver, and terminates an Offboarding vendor with no checklist. Vendors in Offboarding when you reactivate have no checklist and need **Start checklist** (step 13). And when you reactivate, every vendor starts untiered with no assessment history, dependencies or control assessments, no tier requires any control, the default tiers, intake questions, evidence types, contract terms and offboarding items come back, the notification settings return to their defaults (reminders on, automatic reassessment off), and you grant the Vendor Risk Management permissions again. Take a database backup before you deactivate on anything but a test system.

## The API

The screens call the v2 API, so everything in them is available to scripts. Authenticate as you would for any v2 call (see [Authentication and API Keys](https://support.simplerisk.com/kb/08-02-authentication-and-api-keys?hsLang=en); key-based access needs the API Extra). Each user's API permissions match their permissions in the UI.

The vendor record is SimpleRisk's and lives under `/api/v2/vendors`. Every route here needs **Allow Access to "Vendor Risk" Menu**; the table lists what each needs on top.

- **`GET /vendors` (filters: `search`, `status`, `category_id`, `sort` (`name`, `status`, `category` or `updated_at`), `dir`, `start`, `length`), `GET /vendors/{id}`, `GET /vendors/options` (`search`, `limit`), `GET /vendors/lookups`:** Nothing more
- **`POST /vendors`:** Add
- **`POST /vendors/update`, `POST /vendors/status`:** Edit. With this Extra active, a move to Active also needs Approve and a tier, and a move from Offboarding to Terminated also needs Approve and a started checklist with every required step done.
- **`POST /vendors/delete`:** Delete
- **`GET /vendors/{id}/assets`, `POST /vendors/assets/set`:** The Asset Management menu; **Able to Edit Vendors** and **Able to Edit Assets** to save
- **`GET /vendors/{id}/risks`:** The Risk Management menu
- **`GET /assets/{id}/vendors`, `POST /assets/vendors/set`:** The Asset Management menu; **Able to Edit Assets** to save
- **`GET /risks/{id}/vendors`, `POST /risks/vendors/set`:** The Risk Management menu; **Able to Modify Risk Details** to save

Creating an asset with `POST /assets` or a risk with `POST /risks` also accepts the vendors to link, as `vendor_ids[]` together with `vendor_ids_sent=1`. Without the marker the list is ignored, and without the Vendor Risk menu permission both are. On `POST /vendors/assets/set`, a list you leave out is left as it is, and an empty list clears it. A list sends the whole set you want, up to 10,000 ids, and at most 200 of them can be links the vendor doesn't have yet; past either limit the call is refused with a 422 and nothing changes.

This Extra's routes live under `/api/v2/vendorrisk`, and each also needs **Allow Access to "Vendor Risk" Menu**:

#### Register

- **Routes:** `GET /vendorrisk/vendors` (the `/vendors` filters plus `tier_id`, a tier id or `none` for untiered vendors, and `sort` also accepts `tier` and `next_review_date`), `GET /vendorrisk/vendors/profile`, `POST /vendorrisk/vendors/profile` (services and next review date)
- **Permission:** View to read; Manage to write

#### Contacts

- **Routes:** `/vendorrisk/vendors/contacts`, `/vendorrisk/vendors/contacts/update`, `/vendorrisk/vendors/contacts/delete`
- **Permission:** View to read; Manage to write

#### Risk score

- **Routes:** `GET /vendorrisk/vendors/risk` (`vendor_id`, `start`, `length`: the current score with its factors, and the history newest first)
- **Permission:** View together with every permission the score is built from; others get a 403

#### Tiers

- **Routes:** `GET /vendorrisk/tiers` (add `include=version` for the ladder's version), `POST /vendorrisk/tiers/save` (send that `version` back to have the save refused if the ladder changed in the meantime)
- **Permission:** View to read; Configure to write

#### Intake questions

- **Routes:** `GET /vendorrisk/intake/config`, `POST /vendorrisk/intake/questions/save`, `POST /vendorrisk/intake/questions/delete`, `GET /vendorrisk/intake/form`
- **Permission:** Configure for the first three; Manage for the form a requester answers, which carries no points

#### Intakes

- **Routes:** `GET /vendorrisk/intakes`, `GET /vendorrisk/intakes/single`, `POST /vendorrisk/intakes`, `POST /vendorrisk/intakes/decide`
- **Permission:** View to read; Manage to submit (plus Add for a new vendor); Approve together with View to decide

#### Vendor tier

- **Routes:** `POST /vendorrisk/vendors/tier`, `GET /vendorrisk/vendors/tier-history`
- **Permission:** Approve together with View to set; View to read

#### Import

- **Routes:** `GET /vendorrisk/migration/suggestions`, `POST /vendorrisk/migration/apply`
- **Permission:** Configure, plus Manage to apply, plus the Assessments permission; Add to create vendors

#### Pickers

- **Routes:** `GET /vendorrisk/assessment-contact-options`, `GET /vendorrisk/user-options`
- **Permission:** Manage (the contact picker also needs the Assessments permission and the Assessments Extra)

#### Assessment records

- **Routes:** `GET /vendorrisk/diligence` (filters `vendor_id`, `status`, `source`, plus `start` and `length`), `GET /vendorrisk/diligence/single` (`id`), `POST /vendorrisk/diligence`, `POST /vendorrisk/diligence/update`, `POST /vendorrisk/diligence/delete`, `POST /vendorrisk/diligence/sync` (refresh one questionnaire record now)
- **Permission:** Due Diligence together with View

#### Due list

- **Routes:** `GET /vendorrisk/diligence/due` (`within_days`, 0 to 365, defaulting to the review lead time; `state`, `overdue` or `due_soon`, to list only that state, counted before paging; `start`, `length`)
- **Permission:** Due Diligence together with View

#### Sending

- **Routes:** `GET /vendorrisk/diligence/send-options` (`vendor_id`), `POST /vendorrisk/diligence/send` (`vendor_id`, `template_id`, `contact_keys`)
- **Permission:** Due Diligence together with View, plus the Assessments permission; the send also needs Able to Add Questionnaires and Able to Send Questionnaires

#### Tier policies

- **Routes:** `GET /vendorrisk/tier-policies`, `POST /vendorrisk/tier-policies/save` (the whole list of `{tier_id, template_id, required_evidence_type_ids}`: for a caller with the Assessments permission, a tier left out, or sent with a null template, ends up with no template; for anyone else the templates are never changed. A tier or entry that leaves out `required_evidence_type_ids` keeps its list, and an empty list clears it)
- **Permission:** View to read; Configure to save. Templates change only for a caller with the Assessments permission; anyone else's template values are ignored

#### Settings

- **Routes:** `GET /vendorrisk/settings`, `POST /vendorrisk/settings/save` (`review_lead_days`, `notifications_enabled`, `auto_reassess`, `evidence_lead_days`, `contract_lead_days`, `max_file_mb`, `finding_lead_days`, `finding_likelihood_critical`, `finding_likelihood_high`, `finding_likelihood_medium`, `finding_likelihood_low`, `finding_impact_critical`, `finding_impact_high`, `finding_impact_medium`, `finding_impact_low`, `extract_max_calls`, `extract_max_tokens`, `extract_user_hourly`; a setting you leave out is left as it is). Both answers also carry `effective_file_limit_bytes`, the limit in force, `extraction_usage`, the 30-day `runs`, `calls` and `tokens` across the instance, and `ai_provider_configured`, none of which can be saved.
- **Permission:** Configure; turning `auto_reassess` on, and while it's on turning `notifications_enabled` on or raising `review_lead_days`, also need the Assessments permission, Able to Add Questionnaires and Able to Send Questionnaires

#### Evidence types

- **Routes:** `GET /vendorrisk/evidence-types` (`include_disabled=1` adds disabled types and an in-use count, for Configure only), `POST /vendorrisk/evidence-types/save`, `POST /vendorrisk/evidence-types/delete`
- **Permission:** View to read; Configure to write

#### Evidence

- **Routes:** `GET /vendorrisk/evidence` (filters `vendor_id`, `type_id`, `status`, `expiring_within_days`, plus `start` and `length`), `GET /vendorrisk/evidence/single`, `GET /vendorrisk/evidence/gaps` (`vendor_id`), `GET /vendorrisk/evidence/download`
- **Permission:** View, plus Due Diligence or Manage Evidence, to read

#### (unnamed)

- **Routes:** `POST /vendorrisk/evidence`, `POST /vendorrisk/evidence/update` (both accept a `file` as multipart form data), `POST /vendorrisk/evidence/delete`
- **Permission:** View and Manage Evidence

#### Evidence from answers

- **Routes:** `GET /vendorrisk/evidence/importable-files` (`vendor_id`), `POST /vendorrisk/evidence/import` (`vendor_id`, `questionnaire_file_id`, `type_id`)
- **Permission:** View and Manage Evidence, plus the Assessments permission and Able to View Questionnaire Results

#### Contract terms

- **Routes:** `GET /vendorrisk/terms` (the enabled catalog, without the tier requirements)
- **Permission:** View and View Contracts

#### (unnamed)

- **Routes:** `GET /vendorrisk/terms/config` (adds the tier requirements and how many contracts hold a value), `POST /vendorrisk/terms/save`, `POST /vendorrisk/terms/delete`, `POST /vendorrisk/terms/tiers`
- **Permission:** Configure

#### Contract extraction

- **Routes:** `POST /vendorrisk/contracts/extract` (`contract_id`; answers 202 with the run, which is queued), `POST /vendorrisk/contracts/extraction/cancel` (`run_id`; a queued or running run)
- **Permission:** View, View Contracts and Extract

#### (unnamed)

- **Routes:** `GET /vendorrisk/contracts/extraction` (`contract_id`; the newest run, with the requester's name, its status, the reason it failed and its call counts, and the proposals of the newest finished run, each with the contract's current value and that value's `current_fingerprint`)
- **Permission:** View, View Contracts, and Extract or Manage Contracts

#### (unnamed)

- **Routes:** `POST /vendorrisk/contracts/proposals/accept` (`id`, optionally `value` for an edited value, and `confirm_overwrite` with the `current_fingerprint` you read to replace a different stored value; send JSON, because a form can't send the boolean), `POST /vendorrisk/contracts/proposals/reject` (`id`)
- **Permission:** View and Manage Contracts

#### Contracts

- **Routes:** `GET /vendorrisk/contracts` (filters `vendor_id`, `search`, `status`, `expiring_within_days`, plus `start` and `length`; never the value, SLA summary or notes), `GET /vendorrisk/contracts/single`, `GET /vendorrisk/contracts/download`
- **Permission:** View and View Contracts

#### (unnamed)

- **Routes:** `POST /vendorrisk/contracts`, `POST /vendorrisk/contracts/update` (both accept a `file`), `POST /vendorrisk/contracts/delete`, `POST /vendorrisk/contracts/terms` (`id`, and `terms` as a map of term id to value, where an empty or null value clears the term)
- **Permission:** View and Manage Contracts

#### Terms matrix

- **Routes:** `GET /vendorrisk/contracts/matrix` (`term_ids`, `vendor_search`, `tier_id`, `deviations_only`, `include_ended`, `term_filter` with `term_id`, `op` and `value`, `sort` as `vendor`, `tier`, `flags` or `term:{id}`, `dir`, `start`, `length` up to 100; each row carries `flags`, the number of flagged enabled terms, and `deviations_only` keeps the rows where it is above 0)
- **Permission:** View and View Contracts

#### Findings

- **Routes:** `GET /vendorrisk/findings` (filters `vendor_id`, `statuses` (a list, or `all`; open findings when left out), `risk_levels` (a list of level names), `due_state` (`overdue` or `due_soon`), `owner_id` (a user id, or `me`), `source`, `search` on the title, `sort` (`due_date`, `risk_score`, `status`, `vendor`, `title` or `created_at`), `dir`, `start`, `length`; never the description or notes, and a `risk_details` map with the risk's ID and status only where the caller may see them), `GET /vendorrisk/findings/single` (`id`), `GET /vendorrisk/findings/scale` (the likelihood and impact choices and the risk levels)
- **Permission:** View, plus Due Diligence or Manage Findings

#### (unnamed)

- **Routes:** `POST /vendorrisk/findings` (`vendor_id`, `title`, `likelihood`, `impact`, `source`, and optionally `status`, `owner`, `due_date`, `description`, `remediation_notes`, `source_assessment_id`), `POST /vendorrisk/findings/update` (`id` and the fields to change), `POST /vendorrisk/findings/delete` (`id`)
- **Permission:** View and Manage Findings; naming a `source_assessment_id` also needs Due Diligence

#### Findings and risks

- **Routes:** `GET /vendorrisk/findings/risk-options` (`id`: the dialog's defaults and the live likelihood and impact scales), `POST /vendorrisk/findings/risk/create` (`id`, and optionally `subject`, `likelihood`, `impact`)
- **Permission:** View and Manage Findings, plus Able to Submit New Risks and the Risk Management menu

#### (unnamed)

- **Routes:** `POST /vendorrisk/findings/risk/link` (`id`, `risk_id`)
- **Permission:** View and Manage Findings, plus the Risk Management menu and Able to Modify Risk Details, and a risk the caller can see

#### (unnamed)

- **Routes:** `POST /vendorrisk/findings/risk/unlink` (`id`)
- **Permission:** View and Manage Findings

#### Offboarding

- **Routes:** `GET /vendorrisk/offboarding` (`vendor_id`; the open findings count only for a caller who can read findings)
- **Permission:** View

#### (unnamed)

- **Routes:** `POST /vendorrisk/offboarding/start` (`vendor_id`), `POST /vendorrisk/offboarding/step` (`vendor_id`, `step_id`, and `done`, `notes` or both)
- **Permission:** View and Manage

#### Offboarding items

- **Routes:** `GET /vendorrisk/offboarding-items`, `POST /vendorrisk/offboarding-items/save` (`id` to edit, `name`, `description`, `required`, `enabled`, `sort_order`), `POST /vendorrisk/offboarding-items/delete` (`id`)
- **Permission:** Configure

#### Dependencies

- **Routes:** `GET /vendorrisk/dependencies` (`vendor_id`; both directions, with `relied_on_by` capped at 200 rows and `relied_on_by_total` counting them all)
- **Permission:** View

#### (unnamed)

- **Routes:** `POST /vendorrisk/dependencies` (`vendor_id`, `depends_on_vendor_id`, `service`, optionally `notes`), `POST /vendorrisk/dependencies/update` (`id`, and `service`, `notes` or both; the vendors never change), `POST /vendorrisk/dependencies/delete` (`id`)
- **Permission:** View and Manage; both vendors must be visible

#### Required controls

- **Routes:** `GET /vendorrisk/controls/options` (`search` on the control's number or name, `limit` up to 50; Third-Party Management controls first), `GET /vendorrisk/tier-controls` (every tier's required controls, with removed ones marked), `POST /vendorrisk/tier-controls/save` (`tier_id` and `control_ids`, the whole set: an empty list clears it, and a missing one is refused)
- **Permission:** View and Configure, plus **Allow Access to "Governance" Menu**

#### Control coverage

- **Routes:** `GET /vendorrisk/vendors/controls` (`vendor_id`; the exception's ID and name only for a caller with **Able to View Exceptions**), `POST /vendorrisk/vendors/controls` (`vendor_id`, `control_id`, and `status` (`met`, `partial`, `not_met`, or null to clear), `assessed_on`, `notes`, any of them), `POST /vendorrisk/vendors/controls/clear` (`vendor_id`, `control_id`; also for removed controls)
- **Permission:** View and Due Diligence, plus **Allow Access to "Governance" Menu**

#### (unnamed)

- **Routes:** `POST /vendorrisk/vendors/controls/waiver` (`vendor_id`, `control_id`, `exception_id`; null or empty unlinks)
- **Permission:** As above; linking also needs **Able to View Exceptions**

#### (unnamed)

- **Routes:** `GET /vendorrisk/controls/exceptions` (`control_id`; the control's open exceptions, by name)
- **Permission:** As above, plus **Able to View Exceptions**

#### Overview

- **Routes:** `GET /vendorrisk/overview`
- **Permission:** View; each section only for a caller who can read its source

#### Reports

- **Routes:** `GET /vendorrisk/reports/tier-mix` (`statuses`, `tier_ids` (0 for untiered), `search` on the name), `GET /vendorrisk/reports/concentration` (the asset part only with **Allow Access to "Asset Management" Menu**), `GET /vendorrisk/reports/register` (the same filters plus `sort` (any of the caller's columns), `dir`, `start`, `length`; the columns the caller can read, and each row's vendor in `vendor_ids`)
- **Permission:** View

#### (unnamed)

- **Routes:** `GET /vendorrisk/reports/register/export` (`format`, `csv` or `xlsx`, and the register's filters; never its sort or paging)
- **Permission:** View; a file attachment, with only the caller's column groups

#### (unnamed)

- **Routes:** `GET /vendorrisk/reports/heat-map` (the tier-mix filters)
- **Permission:** View and Due Diligence

#### (unnamed)

- **Routes:** `GET /vendorrisk/reports/coverage` (optionally `tier_id`)
- **Permission:** View and Due Diligence, plus **Allow Access to "Governance" Menu**

#### Administration

- **Routes:** `GET /vendorrisk/version`, `POST /vendorrisk/upgrade/db`
- **Permission:** Administrators only

Every route that changes data is a POST, with two exceptions. The Extra's database upgrade route is a GET. And the contract extraction read (`GET /vendorrisk/contracts/extraction`) marks a run whose worker died, and whose lease has run out, as interrupted, which frees the contract for a new request; it changes nothing else. Don't build a monitor or cache that assumes that GET never writes. Responses use the status codes you'd expect: 403 for a missing permission (including a move to Active or to Terminated by someone who can't approve, any evidence route called with only contract permissions, or the reverse, every finding route called without a findings permission, and a waiver link without **Able to View Exceptions**), 404 for a vendor, assessment record, evidence item, contract, file, extraction run, proposal, finding, offboarding step, risk, dependency, control or exception that doesn't exist or isn't visible (a dependency whose other vendor is hidden gets the same 404 as one that doesn't exist, and so does a control that isn't an active SCF control), 409 when a tier, evidence type or contract term is still in use, a term's type or used choices would change, the tier ladder or the tier policies changed since they were read, the Assessments Extra or its contacts aren't available, the SCF Extra isn't available (every required-control and coverage route, **Clear** and the coverage report included), a finding is already linked to a risk or is being sent to the risk register, a finding's likelihood or impact is off the current scale, a checklist can't be started or changed in the vendor's status, an extraction is refused (no AI provider configured, a run already active on the contract or another request of the same user in progress, the document's budget used up, or the document changed while the request was handled), a run has already ended when it's cancelled, or a proposal was already decided, came from a replaced document, or would replace a different stored value without a matching confirmation (the answer then carries the value's `current_fingerprint`), 412 when the encryption key is unavailable, on a save, a download, an extraction request or an accept alike, 422 with a message per field when input is invalid (including a move to Active without a tier, a move to Terminated before the checklist is done, a send while the vendor already has a questionnaire open, a duplicate or 201st dependency, a control the vendor's tier doesn't require, an exception for another control, a tier-controls save naming controls that aren't active SCF controls, a report or export over the 20,000-vendor ceiling or the memory guard, a contract with no document, or one that isn't a PDF or Word file or is too large to read, and an accepted value that doesn't fit the term or date, a null value, a not-found proposal accepted without a value, or a proposal of a term turned off or redefined since), 429 when a user's hourly extraction limit is used up (the answer's `minutes` says when the next request is accepted), 503 when an extraction couldn't be queued (the request still counts against the hourly limit), 500 when a risk was created from a finding but couldn't be linked to it (the response names the risk's ID; link it with `POST /vendorrisk/findings/risk/link`), and 502 when the Assessments Extra couldn't create a questionnaire, in which case nothing was sent and the send can be retried, or when the risk register refused to create a risk from a finding, in which case the finding is unchanged. Template names in the due diligence responses are left out for callers without the Assessments permission. Both download routes and the register export answer with the file as an attachment, never shown inline in the browser, and a refused export answers in JSON rather than with a partial file. The finding reads give a linked risk's ID and status (and, on the single read, its subject) only to a caller with the Risk Management menu who can see the risk; anyone else gets the link's state alone (linked, broken or none). The full request and response shapes, for both families and for the findings, risk, offboarding, dependency, coverage and report routes, are under **vendors** in the in-product API documentation described in [The OpenAPI documentation](https://support.simplerisk.com/kb/08-05-the-openapi-documentation?hsLang=en).

## Common pitfalls

- **Granting Extra permissions without the menu permission.** **Able to View Vendors** on its own shows nothing. Every user of the Extra needs **Allow Access to "Vendor Risk" Menu**.
- **Reading "Able to Manage Vendors" as control of the vendor record.** It manages the Extra's data. Editing, adding and deleting vendors take **Able to Edit Vendors**, **Able to Add Vendors** and **Able to Delete Vendors**.
- **Granting Approve without View.** The approver can't open **Onboard Vendors**, so the permission looks broken. Give **Able to View Vendors**, or a permission that includes it. **Able to Edit Vendors** is needed only to use the **Change Status** card.
- **Granting the extraction permission by habit.** **Able to Extract Terms from Vendor Contracts** sends contract text to your AI provider. Grant it to the people who should make that call, check your provider agreement covers contract documents, and remember that administrators hold it anyway.
- **Setting a generous upload limit and expecting extraction to read every file.** With a 512 MB worker, extraction reads PDF and Word files of up to about 15 MB, whatever the upload limit. Raise the worker's `memory_limit` or tell your users.
- **Restarting workers in the middle of a busy day.** Every running extraction becomes interrupted, costs its user a request and may leave a decrypted temporary file behind. Restart when the queue is quiet.
- **Giving findings managers no risk permissions and expecting Create risk.** The button needs **Able to Submit New Risks** and **Allow Access to "Risk Management" Menu** as well, and **Link existing risk** needs **Able to Modify Risk Details**. A findings manager without them can track findings but never reaches the risk register.
- **Expecting required custom fields on risks created from findings.** A finding sent to the risk register carries only a subject, scoring, owner, teams and one line of notes, and no custom fields. A required custom field doesn't stop **Create risk**: the risk is created without it, and someone completes it on the risk form. Tell your findings managers to do that straight after creating the risk.
- **Changing the Classic scales and forgetting the findings.** A finding whose stored likelihood or impact is no longer on the scale can't be edited or sent to the risk register until someone picks levels that exist. Review the findings list the day you change the scales.
- **Upgrading with vendors in Offboarding and not starting their checklists.** They can't be terminated until someone uses **Start checklist**. Do it as part of the upgrade.
- **Giving the security team contract access by habit.** Evidence and contract permissions are separate so that the people who read SOC 2 reports don't automatically read pricing and liability caps. Grant **Able to View Vendor Contracts** to the people who own the commercial relationship, and think twice before adding it to an analyst role.
- **Granting Able to Run Vendor Due Diligence on its own.** It needs **Able to View Vendors**, or a permission that includes it, like Approve does. Analysts who should send questionnaires need **Allow Access to "Assessments" Menu**, **Able to Add Questionnaires** and **Able to Send Questionnaires** too. The Assessments menu permission on its own isn't enough.
- **Switching on automatic reassessment with a backlog.** The next daily run sends a questionnaire to every eligible vendor that's due or overdue. Clear the overdue list on the **Conduct Due Diligence** page first, or accept the burst.
- **Leaving business owners blank.** Review and evidence reminders go only to a vendor's business owner, and contract reminders go to it when the contract has no owner. A vendor without one, or whose owner can't see it, is never reminded about.
- **Leaving a contract owner who can't see contracts.** The reminder goes nowhere, and it doesn't fall back to the business owner. Check that contract owners hold **Able to View Vendor Contracts**.
- **Setting a 25 MB limit on a stock database.** The card will tell you files up to about 3.5 MB can be uploaded, and that's the number your users get. Raise the MySQL and PHP limits if you need bigger files.
- **Turning on Assessments' own scheduling for due diligence questionnaires.** Their reminder and recurring-send settings are off on purpose. Switching on a recurring send there runs a second cycle next to the tier's interval.
- **Expecting teams to hide vendors without the Team-Based Separation Extra.** They don't. If vendor details are sensitive, turn the Extra on before you start loading vendors.
- **Leaving vendors without a team under separation.** A vendor with no team is visible to everyone who can open Manage Vendors, which is the opposite of what people expect from a team setting. Administrators can create such vendors; non-administrators can't.
- **Giving requesters Approve.** The person who fills in the questionnaire shouldn't also confirm the tier. Keep the two permissions on different roles.
- **Retuning thresholds and expecting vendors to move.** They don't. Changing a minimum score, a weight or a question affects the next intake only, so a vendor whose tier you now disagree with needs an approver to set its tier.
- **Leaving existing Active vendors untiered.** They keep working, but they show as **Untiered** and can't return from Offboarding to Active until someone sets a tier. An approver can clear the backlog with **Set tier**.
- **Deactivating to "reset" the Extra.** It deletes every tier, intake, contact, risk score, assessment record, evidence file, contract, finding, offboarding checklist, dependency and control assessment, and Active and Terminated stop needing an approver. To clear a handful of vendors, delete them from **Manage Vendors**.
- **Activating on a system with no encryption key, once built-in encryption is available.** It works until someone types a contact's phone number. Check the key directory first.
- **Granting due diligence and expecting a Controls tab.** Coverage also needs **Allow Access to "Governance" Menu**, and choosing required controls needs it on top of **Able to Configure Vendor Risk Management**. There's no locked tab to hint at what's missing.
- **Requiring the whole Third-Party Management family on day one.** Every required control is a row on every vendor in the tier. Start small and add controls as your analysts actually assess them.
- **Uninstalling the SCF Extra casually.** Every requirement and every assessment becomes **Control removed** for good, and reinstalling doesn't bring them back. Export the register first if you want a record of where coverage stood.
- **Approving exceptions without thinking about vendors.** A control exception approved in Governance waives that control for every vendor that links it, and keeps waiving past its review date while it stays approved and open.
- **Assuming the Overview scales without limit.** Above 20,000 vendors the user can see, it's refused as a whole, and there's no filter to narrow it. Raise `memory_limit` if you're near the memory guard, and narrow the register and export with their filters.

## Related

- [Permission Reference](https://support.simplerisk.com/kb/03-03-permission-reference?hsLang=en)
- [Team-Based Separation](https://support.simplerisk.com/kb/03-06-team-based-segregation?hsLang=en)
- [Managing Dropdown Values](https://support.simplerisk.com/kb/05-04-managing-dropdown-values?hsLang=en)
- [Key Management and Rotation](https://support.simplerisk.com/kb/09-03-key-management-and-rotation?hsLang=en)
- [File Upload Size Limits](https://support.simplerisk.com/kb/12-07-file-upload-size-limits?hsLang=en)
- [API Overview](https://support.simplerisk.com/kb/08-01-api-overview?hsLang=en)
- [Managing Vendors in SimpleRisk](https://support.simplerisk.com/kb/04-05-managing-vendors-in-simplerisk?hsLang=en) in the User Guide
- [Using the Vendor Register](https://support.simplerisk.com/kb/06-05-the-vendor-register?hsLang=en) in the User Guide
- [Vendor Intake and Tiering](https://support.simplerisk.com/kb/06-06-vendor-intake-and-tiering?hsLang=en) in the User Guide
- [Vendor Due Diligence and Reassessment](https://support.simplerisk.com/kb/06-07-vendor-due-diligence?hsLang=en) in the User Guide
- [Vendor Evidence and Contracts](https://support.simplerisk.com/kb/06-08-vendor-evidence-and-contracts?hsLang=en) in the User Guide
- [Vendor Findings and Offboarding](https://support.simplerisk.com/kb/06-09-vendor-findings-and-offboarding?hsLang=en) in the User Guide
- [Extracting Contract Terms with AI](https://support.simplerisk.com/kb/06-11-vendor-contract-term-extraction?hsLang=en) in the User Guide
- [Configuring AI Providers](https://support.simplerisk.com/kb/10-05-configuring-ai-providers?hsLang=en)
- [Vendor Overview, Reports and Control Coverage](https://support.simplerisk.com/kb/06-10-vendor-overview-reports-and-control-coverage?hsLang=en) in the User Guide
- [The Secure Controls Framework (SCF) Extra](https://support.simplerisk.com/kb/06-06-configuring-the-complianceforgescf-extra?hsLang=en)
- [The Cron Jobs](https://support.simplerisk.com/kb/02-07-the-cron-jobs?hsLang=en)

- [FAQs](https://support.simplerisk.com/kb/faqs?hsLang=en)
- [SimpleRisk Extras](https://support.simplerisk.com/kb/simplerisk-extras?hsLang=en#main-content)

    - [Vulnerability Management Extra](https://support.simplerisk.com/kb/simplerisk-extras?hsLang=en#vulnerability-management-extra)
    - [Team Separation Extra](https://support.simplerisk.com/kb/simplerisk-extras?hsLang=en#team-separation-extra)
    - [Import-Export Extra](https://support.simplerisk.com/kb/simplerisk-extras?hsLang=en#import-export-extra)
- [Administrator Guide](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#main-content)

    - [00 About This Guide](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#00-about-this-guide)
    - [01 Installation and Deployment](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#01-installation-and-deployment)
    - [02 Upgrades and Maintenance](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#02-upgrades-and-maintenance)
    - [03 Users and Permissions](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#03-users-and-permissions)
    - [04 Authentication](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#04-authentication)
    - [05 Customization](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#05-customization)
    - [06 Configuring Risk and Compliance](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#06-configuring-risk-and-compliance)
    - [07 Integrations](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#07-integrations)
    - [08 The API](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#08-the-api)
    - [09 Encryption and Data Security](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#09-encryption-and-data-security)
    - [10 Workflows and Automation](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#10-workflows-and-automation)
    - [11 Reporting and Auditing](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#11-reporting-and-auditing)
    - [12 Operations](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#12-operations)
    - [13 Reference](https://support.simplerisk.com/kb/administrator-guide?hsLang=en#13-reference)
- [User Guide](https://support.simplerisk.com/kb/user-guide?hsLang=en#main-content)

    - [00 Foundations of GRC](https://support.simplerisk.com/kb/user-guide?hsLang=en#00-foundations-of-grc)
    - [01 Risk Management](https://support.simplerisk.com/kb/user-guide?hsLang=en#01-risk-management)
    - [02 Compliance Management](https://support.simplerisk.com/kb/user-guide?hsLang=en#02-compliance-management)
    - [03 Governance](https://support.simplerisk.com/kb/user-guide?hsLang=en#03-governance)
    - [04 Asset and Data Inventory](https://support.simplerisk.com/kb/user-guide?hsLang=en#04-asset-and-data-inventory)
    - [05 Threat and Vulnerability Management](https://support.simplerisk.com/kb/user-guide?hsLang=en#05-threat-and-vulnerability-management)
    - [06 Assessments](https://support.simplerisk.com/kb/user-guide?hsLang=en#06-assessments)
    - [07 Incident Management](https://support.simplerisk.com/kb/user-guide?hsLang=en#07-incident-management)
    - [08 Audit and Reporting](https://support.simplerisk.com/kb/user-guide?hsLang=en#08-audit-and-reporting)
    - [09 Day-to-Day SimpleRisk](https://support.simplerisk.com/kb/user-guide?hsLang=en#09-day-to-day-simplerisk)
    - [10 Continuous Improvement](https://support.simplerisk.com/kb/user-guide?hsLang=en#10-continuous-improvement)
    - [11 Business Continuity and Disaster Recovery](https://support.simplerisk.com/kb/user-guide?hsLang=en#11-business-continuity-and-disaster-recovery)
- [SimpleRisk User Guides](https://support.simplerisk.com/kb/simplerisk-user-guides?hsLang=en)
- [Troubleshooting](https://support.simplerisk.com/kb/troubleshooting?hsLang=en)
- [SimpleRisk Hosted](https://support.simplerisk.com/kb/simplerisk-hosted?hsLang=en)
- [How To videos](https://support.simplerisk.com/kb/how-to-videos?hsLang=en)

[![favicon-1](https://support.simplerisk.com/hs-fs/hubfs/favicon-1.png?width=35&height=35&name=favicon-1.png "favicon-1")](https://www.simplerisk.com)

<https://www.facebook.com/simplerisk/> <https://www.twitter.com/simpleriskfree/> <https://www.linkedin.com/company/simplerisk/>

Copyright © 2026, SimpleRisk, Inc.