Reviewed 11 October 2026. This is a documentation review, not a test. The award record for 2026 is on the award page.
Pages the panel read
- Relevance AI home
- Relevance AI pricing
- Relevance AI product
- Relevance AI integrations
- Relevance AI API page
- Relevance AI docs
- Relevance AI docs: integrations
Scores
| Dimension | Score | Checks |
|---|---|---|
| Tool Access | 70 of 100 | 2 documented, 3 mentioned, 0 not found |
| Task Setup | 100 of 100 | 5 documented, 0 mentioned, 0 not found |
| Running and Oversight | 80 of 100 | 3 documented, 2 mentioned, 0 not found |
| Billing Clarity | 80 of 100 | 3 documented, 2 mentioned, 0 not found |
The total is the plain sum of the four dimensions, out of 400. Each check is worth 20 points when documented in detail, 10 when mentioned only, and 0 when nothing was found.
Checklist
Tool Access: 70 of 100
How many outside tools the builder connects to, and whether its public pages name them.
| Check | Finding | What the documentation says | Source |
|---|---|---|---|
| Names specific third-party integrations on a public page | Documented | The integrations directory names HubSpot, Slack, Gmail and Notion among many apps. | Relevance AI integrations |
| Connects search engines or SEO data | Mentioned | The docs use a Google search action as an example. No SEO data tool is named. | Relevance AI docs: integrations |
| Connects social or community platforms | Mentioned | Slack and WhatsApp Business are named as messaging connections only. | Relevance AI docs: integrations |
| Connects sales or contact data | Documented | HubSpot contact and deal actions are described. Lusha is named for contact data. | Relevance AI docs: integrations |
| Connects media generation (image, video or voice) | Mentioned | A phone agent is named for voice calls. No image or video generator is named. | Relevance AI API page |
Task Setup: 100 of 100
How a person turns a goal into a working agent without wiring each tool by hand.
| Check | Finding | What the documentation says | Source |
|---|---|---|---|
| Starts from a plain-language task description | Documented | Agents are built from a plain-language description, with tools and tests generated from it. | Relevance AI product |
| Picks or connects tools without manual wiring | Documented | The product page describes building the tools from a description. | Relevance AI product |
| Offers ready-made templates or workflows | Documented | The marketplace offers pre-built agents, tools and workforces to clone. | Relevance AI docs |
| Has a visual, no-code builder | Documented | A visual drag-and-drop builder with a canvas is described for agents and workforces. | Relevance AI product |
| Connects outside AI agents over MCP, SDK or API | Documented | Claude Code, Codex and Cursor can connect through MCP, and an API is documented. | Relevance AI API page |
Running and Oversight: 80 of 100
Whether the agent keeps working on a schedule, asks a person before outside actions, and shows what it did.
| Check | Finding | What the documentation says | Source |
|---|---|---|---|
| Runs on a schedule or continues in the cloud | Documented | Schedules, webhooks and a managed queue for runs are described. | Relevance AI product |
| Asks a person to approve before outside actions | Documented | Human-in-the-loop approvals with per-tool approval rules are described. | Relevance AI product |
| Keeps a run history the user can open | Mentioned | Full agent tracing and audit logs are named. The user-facing log view is not explained. | Relevance AI product |
| Shows which tool or step was used, and why | Mentioned | Tracing is named. A step-by-step view is not explained. | Relevance AI product |
| Supports team sharing or several users | Documented | The pricing page lists unlimited users, role-based access and single sign-on for the enterprise plan. | Relevance AI pricing |
Billing Clarity: 80 of 100
Whether the free start, the price and the usage are stated in public.
| Check | Finding | What the documentation says | Source |
|---|---|---|---|
| States a free way to start | Documented | A free plan is referenced with a try-for-free action. | Relevance AI API page |
| Shows plans or rates on a public page | Mentioned | Only an enterprise plan is shown on the pricing page, without prices. Sales contact is the route. | Relevance AI pricing |
| Uses one balance or one bill across the tools | Mentioned | The enterprise plan mentions vendor credits, with no explanation of how they work. | Relevance AI pricing |
| Shows how much credit or usage each task used | Documented | The dashboard shows tasks, spend and average cost per task. | Relevance AI product |
| Says no credit card is needed to start | Documented | The free plan is described with no card required. | Relevance AI API page |
What the documentation does not cover
The panel found no gaps in this product’s public documentation on the checks above.
Where the panel stopped
The panel read public documentation only. It did not buy the product, run a task, or compare real results. Documentation is not performance: a product may do things it does not publish, and a product that performs well but publishes little will score low here. That limit is stated on every research file.
Compare this file with the ranking page, where the same scores appear. Any difference between the two is an error, and should be reported through the corrections log.
Published 2026‑10‑11 · Panel review of public documentation, October 2026