# Relevance AI — AI Agents & Assistants

Relevance AI by OnSearch Pty Ltd holds a panel rating of 3/5 from 6 AI judges (category median 3/5), as of 2026-10-28. Provenance: Rest of world; sovereignty: 1 of 4 dimensions proven.

- Vendor: OnSearch Pty Ltd (https://relevanceai.com)
- Category: [AI Agents & Assistants](https://whats-best.ai/software/ai-agents.md)
- Provenance (editorial classification): Rest of world
- Canonical page: https://whats-best.ai/software/ai-agents/relevance-ai
- Alternatives: https://whats-best.ai/software/ai-agents/relevance-ai/alternatives ([markdown](https://whats-best.ai/software/ai-agents/relevance-ai/alternatives.md))
- Data as of: 2026-10-29 · evaluated 2026-10-28 · oldest source captured 2026-10-26

**AI-generated.** Every score, rationale and verdict here is written by a panel of AI judges from public evidence — an analysis, not a hands-on test. Every fact quotes the vendor page it was read from. No score can be bought. How scores are made: https://whats-best.ai/methodology

## At a glance

- Vendor: OnSearch Pty Ltd
- Headquarters: Not stated on the vendor’s pages as of 26 Oct 2026
- Data hosting: Not stated on the vendor’s pages as of 26 Oct 2026
- Free plan: Not stated on the vendor’s pages as of 26 Oct 2026

## The panel's verdict

Relevance AI, a no-code platform for building AI agent teams from the Australian company OnSearch Pty Ltd, is judged strongest on agent capabilities, which scores 8: evaluations gate every build path, human-in-the-loop approvals sit on the visual canvas, and the vendor states its own limit that vector search "always returns results, even if none are truly relevant". Admin and access scores 7 and integrations and knowledge 6-7, on single sign-on, role-based access, audit exports to the customer's own S3 bucket, and read-and-write SharePoint and Confluence connectors. The split is on processing and models, spanning 4 to 7, the rest at 6: the skeptic scored 4 because no page pins inference regions or EU-only processing, while the automation lead's 7 credits named model families, routing to any provider, and OpenAI, LLaMA and Fireworks hosted in the vendor's own AWS and Azure. The weakest marks fall on sovereignty at 4-5 — the Australian entity under New South Wales law, prompts to the US providers OpenAI and Anthropic, EU/UK storage selectable at signup — and pricing transparency at 2-3, with no price figures on the captured pages.

**Speaks for it:**
- Agent capabilities scores 8, with evaluations gating deployment, A/B testing, live-run sampling and drift flagging documented.
- SharePoint and Confluence connectors document read-and-write actions and event triggers, alongside MCP, custom connectors, an API and workspace-level shared connections.
- Single sign-on via SAML 2.0 and OIDC, identity-provider group sync, role-based access at organization, project and asset level, and audit events exportable to the customer's own S3 bucket are documented.
- Model families are named — Gemini, GPT, Claude, Kimi and GLM — with routing to any provider, bring-your-own API keys, and OpenAI, LLaMA and Fireworks hosted in the vendor's own AWS and Azure environments.
- The security documentation states no training on customer data without a partnership agreement, a vendor table shows OpenAI and Anthropic with no logging and a DPA, and SOC 2 Type II is claimed.

**Held against it:**
- Sovereignty scores 4-5, with the contracting entity OnSearch Pty Ltd under New South Wales law and prompts reaching the US model providers OpenAI and Anthropic.
- The captured pages give different figures for the integration count — 1,000+, 1,800+ and 2,000+.
- The captured pages give different positions on improvement use: the terms grant a licence to use Client Data to improve the Solution, while the security pages state no training or improvement without a partnership agreement.
- Data retention configuration is Enterprise-only, must be enabled by the vendor's team, and does not reach knowledge bases, integrations or API keys.
- Single sign-on is Enterprise-only, one identity provider per organization is the limit, and an email change provisions a fresh account whose permissions and owned assets do not transfer.

**Best for:**
- You are deploying agents that must write into connected systems — creating, updating or deleting SharePoint and Confluence items — with human approval steps on the canvas.
- You want releases gated by evidence: pre-deployment scenarios, A/B testing, live-run sampling and drift flagging before company-wide rollout.
- Your IT team needs enterprise identity controls — SAML/OIDC single sign-on, identity-provider group sync, and audit logs exportable to your own S3 bucket.
- You want model flexibility — named families, routing to any provider and bring-your-own API keys.

**Avoid if:**
- You require a European contracting entity — the vendor is OnSearch Pty Ltd under New South Wales law, and prompts reach the US model providers OpenAI and Anthropic.
- You need published price figures before talking to sales — the captured pricing pages show no per-tier numbers and single sign-on sits behind an Enterprise contact.
- Your users change email addresses routinely — email is the account identifier and a change provisions a new account whose permissions and owned assets do not transfer.
- You need configurable data retention below the Enterprise tier — it is Enterprise-only and requires the vendor's team to enable it.

## Scores by judge and criterion

Scores out of 5, each judge scoring independently against the published rubric (v1). "n/r" = not rated: the vendor publishes no price, or the judge found no public information on that criterion.

| Criterion | The Head of IT | The Data Protection Officer | The Automation Lead | The Customer-Service Director | The Works Council Member | The Skeptic |
| --- | --- | --- | --- | --- | --- | --- |
| Data use, training & retention | 3.5/5 | 3.5/5 | 3.5/5 | 3.5/5 | 3/5 | 3/5 |
| Models & processing location | 3/5 | 3/5 | 3.5/5 | 3/5 | 3/5 | 2/5 |
| Admin, identity & access controls | 3.5/5 | 3.5/5 | 3.5/5 | 3.5/5 | 3.5/5 | 3.5/5 |
| Integrations, connectors & knowledge | 3.5/5 | 3.5/5 | 3.5/5 | 3/5 | 3.5/5 | 3.5/5 |
| Agent capabilities & guardrails | 4/5 | 4/5 | 4/5 | 4/5 | 4/5 | 4/5 |
| European sovereignty | 2/5 | 2.5/5 | 2/5 | 2/5 | 2/5 | 2/5 |
| Pricing transparency | n/r | n/r | n/r | n/r | n/r | n/r |
| **Weighted verdict** | **3.5/5** | **3/5** | **3.5/5** | **3.5/5** | **3/5** | **3/5** |

### The judges

- **The Head of IT** (v1.0): Has to roll an AI assistant out to eight hundred staff without a second shadow-IT wave. Wants SSO and SCIM, groups that decide who gets which model, agent and connector, answers that respect SharePoint permissions, audit logs for the SIEM, and a deployment option that is not "trust us".
- **The Data Protection Officer** (v1.0): Signs the DPA and gets asked where a prompt goes. Wants every model provider named as a subprocessor with its region, training excluded for all of them, retention the company controls, and an EU-only switch that also covers agents and connectors — not just the chat window.
- **The Automation Lead** (v1.0): Builds the agents that file tickets, draft offers and chase approvals across the ERP, CRM and wiki. Wants tools and actions with scoped permissions, MCP and an API, approval steps before anything is sent, run logs, and a way to test an agent before the whole company uses it.
- **The Customer-Service Director** (v1.0): Runs a contact centre with forty agents and wants half the routine calls and chats answered by an AI agent. Wants handover to a human with the full context, actions in the order and ticket systems, evaluation before go-live, and a price per conversation the CFO can forecast.
- **The Works Council Member** (v1.0): Has a co-determination right the moment a tool can monitor staff, and an AI assistant logs every prompt. Wants usage reports that can be aggregated or pseudonymised, prompt logging rules set out in writing, a human approving what agents do to colleagues, and no training on staff input.
- **The Skeptic** (v1.0): Has read "European AI" on a wrapper that sends every prompt to a US model, "agents" that are saved prompts, and "we never train on your data" with no word on the model provider behind it. Asks which model answers, where it runs, what it may do unsupervised, and what a thousand seats cost.

## European sovereignty — proven facts

Built only from facts on the vendor's own pages; a dimension we could not prove is left open, never scored as zero. Headline: 1 of 4 dimensions proven.

| Dimension | Finding | Points | Source |
|---|---|---|---|
| Legal entity | Not determined | — | — |
| Ownership | Not determined | — | — |
| Data residency | Not determined | — | — |
| Subprocessors | US CLOUD Act reach | 0/2 | [S9] |

## Key facts

Each fact is the vendor's own statement, cited to the source page it was read from ([S*n*] → Sources).

**Compliance**
- Audit logs: yes [S13]
- Australia rights: yes [S6]
- Background checks: employees with access to Subscriber Data · yes [S4]
- Children privacy: no [S6]
- Customer managed keys: yes [S9]
- Data privacy features: yes · yes · yes · yes [S1]
- Data residency: yes [S13]
- Data retention: as long as necessary to provide Services or for marketing; longer for legal obligations [S6]
- Deidentified data: yes · no [S6]
- DPA reference: yes [S9]
- Encryption: AES-256 [S4]
- Encryption: most information · yes [S6]
- Encryption: AES 256 · TLS 1.2+ [S9]
- GDPR: yes [S1]
- GDPR: yes [S9]
- GDPR: yes [S13]
- GDPR rights: privacy@relevanceai.com · yes [S6]
- GDPR support: yes [S7]
- GDPR transfer mechanisms: yes · yes · yes [S6]
- Industry regulations: SOX · PCI-DSS · eDiscovery · FISMA [S7]
- No training on data: yes [S13]
- Penetration testing: written agreement · no [S4]
- Pii masking: yes [S13]
- Rbac: yes [S13]
- Security questionnaire: enterprise · yes · yes [S9]
- Security training: yes · yes [S4]
- Service provider role: service provider / processor on behalf of Customers [S6]
- Slack app data handling: yes · no [S6]
- SOC 2: yes [S1]
- SOC 2: AICPA SOC 2 Type II [S13]
- SOC 2 type2: yes [S9]
- SSO rbac audit logs: yes · yes · yes [S3]
- SSO SAML: yes [S13]
- Third party assessments: enterprise · yes · yes [S9]
- Third party liability: no [S4]
- Trust center: Trust Center listed [S14]

**Data**
- Activity definition: manual runs · automated or scheduled runs [S8]
- Client data licence: non-exclusive, royalty-free, non-transferable, worldwide, irrevocable, for providing and improving the Solution [S5]
- Deletion permanent: no [S8]
- Deletion permanent: no · no [S7]
- Deletion scope: yes · no [S8]
- Deletion SLA: 60 [S9]
- Enterprise exemption: no · yes [S8]
- Enterprise retention: enterprise · yes · yes [S9]
- Export formats: CSV · Excel · JSON [S9]
- Hosting: possibly outside Australia unless otherwise agreed · third-party hosting service selected by vendor [S5]
- Inactive table deletion event: 2026-05-04 · Knowledge tables on inactive accounts · yes [S8]
- Inactivity threshold: 182 · yes · yes · yes [S8]
- Model training: no [S9]
- Ownership: yes [S9]
- Prevent deletion: login or run any Agent/Tool [S8]
- Residency regions: US (N. Virginia) · EU (London) · AU (Sydney) · yes [S9]
- Retention disable toggle: yes · yes [S7]
- Retention enterprise only: enterprise_only · yes [S7]
- Retention exclusions: knowledge bases · integrations · API keys · organization/project settings [S7]
- Retention export before deletion: yes [S7]
- Retention logs: until deleted · 30 [S9]
- Retention no notifications: no [S7]
- Retention on cancellation: until end of current billing cycle [S5]
- Retention org wide: no [S7]
- Retention period configurable days: days · 30 · 90 · 180 · 365 [S7]
- Retention s3 streaming exempt: no [S7]
- Retention scope: tool runs & sync items · agent conversations · workforce tasks · agent/tool/workforce versions [S7]
- Retention staging bucket purge: 30 [S7]
- Retention subprocessor out of scope: no [S7]
- Retention version protection: active versions · draft versions [S7]
- Retention who configures: organization admins · owners [S7]
- Separate from enterprise retention: yes · Enterprise-only configuration for lifecycle of tool runs, agent conversations, and activity data [S8]
- Subject rights: yes · yes · yes [S6]
- Warning email: support@relevanceai.com · 4 [S8]

**Hosting**
- Iaas provider: unnamed cloud infrastructure provider hosts data in data centre facilities [S4]
- Region: data centres may be located in various global regions [S4]
- Tenant isolation: multi-tenant, logical separation [S9]

**Integrations**
- API keys: yes · yes [S5]
- API mcp: yes · yes [S3]
- API mcp: API & MCP offered [S14]
- App count: 1,000+ native connectors [S13]
- Apps: 1000+ [S1]
- Count 2000: 2000 · >= [S3]
- Custom actions gpts: Custom Actions for GPTs listed [S14]
- Custom connectors: Custom connectors supported [S13]
- Otel delta share: OTEL and Delta Share export [S13]
- Pages: yes [S1]
- Sharepoint actions: 4 · Download File · Update Item · Create List · Create Item [S12]
- Sharepoint triggers: 2 · Updated List Item · New List Item [S12]
- Slack app: yes [S6]
- SSO identity providers: WorkOS · Okta · Azure AD / Microsoft Entra ID · Google Workspace · Auth0 [S10]
- Total integrations: 1800 [S11]
- Total integrations: 1800 [S12]
- Triggers: Native CRM/email/calendar triggers plus webhooks, cron, custom triggers [S13]
- Workspace shared connection: yes [S12]

**Legal**
- Copyright footer: 2026 · OnSearch Pty Ltd T/A Relevance AI [S5]
- Entity: OnSearch Pty Ltd · Relevance AI [S3]
- Entity: OnSearch Pty Ltd T/A Relevance AI [S1]
- Entity: Relevance AI · OnSearch Pty Ltd. [S6]
- Entity: 33 637 909 409 · Australia · Relevance AI · OnSearch Pty Ltd [S5]
- Entity: OnSearch Pty Ltd T/A Relevance AI [S14]
- Entity: OnSearch Pty Ltd T/A Relevance AI [S13]
- Governing law: New South Wales, Australia · courts of New South Wales [S5]
- Liability exclusion: yes [S4]
- Operating entity: OnSearch Pty Ltd T/A Relevance AI [S4]
- Operating entity: OnSearch Pty Ltd · Relevance AI [S12]
- Policies: Privacy policy, Security policy, Terms listed [S14]
- Privacy policy: https://relevanceai.com/privacy-policy/ [S5]
- Terms update: yes [S5]

**Pricing**
- Billing renewal: yes [S5]
- Card surcharges: Visa · MasterCard · American Express · yes [S5]
- Custom vendor credits: Custom Vendor Credits [S3]
- Fee components: Subscription Fees · Actions · Vendor Credits [S5]
- Free account: yes [S11]
- Free account: yes [S12]
- Gst: GST · no [S5]
- Liability cap: fees paid in the 3 months preceding the liability event [S5]
- Model routing: Routing to any model provider supported [S13]
- Refund policy: no [S5]
- Unlimited users projects: yes · yes [S3]
- Vendor credits markup: no [S5]

**Product**
- Ab testing analytics: A/B Testing & Analytics [S3]
- Access control: yes · yes [S4]
- Access controls: yes · yes · yes · yes [S1]
- Agent evaluations: Agent Evaluations [S3]
- Agent models: Agents run on multiple models including Gemini 3 Flash, GPT-5.4 Mini, Kimi K2.6, Claude Haiku 4.5, Gemini 3.1 Pro, GLM-5.2, Claude Sonnet 4.6 [S1]
- Agent tracing: yes [S13]
- AI agent security: yes · yes [S9]
- Audit events: S3 [S9]
- Backup dr: automatic, encrypted, multi-AZ · yes [S9]
- Build methods: 3 · drag and drop visual canvas · build with AI (Invent) · build with MCP [S13]
- Calling meeting agents: Calling & Meeting Agents [S3]
- Confluence actions: Update a Post, Search Content (CQL), Delete Post (irreversible), Create Post, Create Page [S11]
- Confluence integration: Confluence integration allows creating, updating, deleting content, spaces, and more [S11]
- Confluence tool steps: 5 · 3 [S11]
- Confluence triggers: Watch Pages, Watch Blog Posts, New Page or Blog Post [S11]
- Copilot cowork: Copilot product named Cowork for research and drafting [S1]
- Cost visibility: yes [S13]
- Custom actions: Custom Actions [S3]
- Data masking: visual data masking · yes [S9]
- Departments: Use cases for Sales, Customer Success, Marketing, Human Resources (+3 more) [S1]
- Deployment onboarding: Embedded deployment team for onboarding [S13]
- Deployment timeline: Embedded deployment team plus onboarding timeline weeks 1-2 mapping use cases, week 3+ first agent team, week 4+ production [S2]
- Disaster recovery: yes · Service Level Policy · Service Level Policy [S4]
- Drag drop canvas: yes [S2]
- Employee data access: no [S4]
- Enterprise: Enterprise offering with access controls, audit trails, and residency guarantees [S1]
- Enterprise feature list: S3 Event Streaming · Role-based access controls · Analytics · Integration controls [S7]
- Enterprise fga isolation: enterprise · yes [S9]
- Enterprise triggers: Enterprise Triggers [S3]
- Event streaming: OpenTelemetry · yes [S10]
- Firewalls: network · host · application [S4]
- Functions: Functions: Sales, Marketing, Operations, Research, Support [S14]
- Hallucination reduction: Multi-step process reduces risk of irrelevant or hallucinated responses [S14]
- Human in loop approvals: yes [S13]
- Human oversight: yes [S2]
- Human oversight: Oversight of tasks, decisions and escalations in one place [S13]
- Invent: Invent 2.0, a 24/7 AI solutions engineer that helps identify what to automate, turn expertise into reliable systems, and improve systems over time [S1]
- Knowledge search step: Relevance provides a knowledge search step in its tool builder [S14]
- L4 autonomy: L4: agents run evals and swap models themselves [S1]
- Levels: Platform maps path from assisted AI (L1) to full autonomy (L4): L1 Assisted, L2 Copilot, L3 Autopilot, L4 Self-Driving [S1]
- Live run sampling: yes [S2]
- Live run sampling: Sampling of live runs with pass-rate charting and drift flagging [S13]
- Llm vendor table: OpenAI · Anthropic · no · yes · no [S9]
- Min eval bar: 0.9 [S1]
- Model routing: Route to any model provider, optimize cost per agent and per task [S2]

**Subprocessors**
- List URL: https://trust.relevanceai.com/subprocessors [S9]
- Regions: United Kingdom · European Economic Area · United States · Australia [S6]

**Support**
- Dedicated account manager: yes [S3]
- Docs: yes · yes · yes [S1]
- Docs channels: Support resources: Documentation, Changelog, API & MCP [S14]
- Documentation: yes [S3]
- Documentation: contact support · documentation · changelog · API & MCP [S13]
- Enterprise dedicated: enterprise · yes [S7]
- Privacy contact: privacy@relevanceai.com [S6]
- SSO setup time: 30 · 60 [S10]
- Warning email dispute: reply to warning email to open a support ticket [S8]

42 more facts on the canonical page: https://whats-best.ai/software/ai-agents/relevance-ai

## Questions

### Is Relevance AI European?

No. whats-best.ai classifies Relevance AI as Rest of world.

### Where does Relevance AI store data?

Not stated on the vendor’s pages as of 26 Oct 2026.

### Does Relevance AI have a free plan?

Not stated on the vendor’s pages as of 26 Oct 2026.

### What are the best alternatives to Relevance AI?

As of 28 Oct 2026, the panel rates [Langdock (4/5)](https://whats-best.ai/software/ai-agents/langdock), [Dust (3.5/5)](https://whats-best.ai/software/ai-agents/dust) and [Unique AI (3.5/5)](https://whats-best.ai/software/ai-agents/unique-ai) highest among the other 15 AI Agents & Assistants products evaluated. Ratings that look equal are rounded to half stars; the order uses the unrounded rating, then the name. The highest-rated European alternative is [Langdock](https://whats-best.ai/software/ai-agents/langdock) (EU-Made). Ratings are the AI panel’s opinion, formed from public evidence: [how scores are made](https://whats-best.ai/methodology#ordering). The full list, with European products first: [Relevance AI alternatives](https://whats-best.ai/software/ai-agents/relevance-ai/alternatives).

## What this page leaves out

- 6 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 compliance facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 legal facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 pricing facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 1 integrations fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.
- 1 support fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.
- 1 sovereignty dimension could not be confirmed on the vendor’s own pages and is shown as unknown.

## Sources

- [S1] Vendor homepage: https://relevanceai.com/ (checked 2026-10-26)
- [S2] Features page (agents): https://relevanceai.com/agents (checked 2026-10-26)
- [S3] Pricing page: https://relevanceai.com/pricing (checked 2026-10-26)
- [S4] Data security policy: https://relevanceai.com/data-security-policy (checked 2026-10-26)
- [S5] Terms naming the legal entity: https://relevanceai.com/terms-and-conditions (checked 2026-10-26)
- [S6] Privacy policy: https://relevanceai.com/privacy-policy (checked 2026-10-26)
- [S7] Data use, training & retention — found from sitemap: https://relevanceai.com/docs/enterprise/data-retention (checked 2026-10-26)
- [S8] Data use, training & retention — found from sitemap: https://relevanceai.com/docs/build/knowledge/inactive-table-deletion (checked 2026-10-26)
- [S9] Admin, identity & access controls — found from sitemap: https://relevanceai.com/docs/admin/security (checked 2026-10-26)
- [S10] Admin, identity & access controls — found from sitemap: https://relevanceai.com/docs/enterprise/sso-setup (checked 2026-10-26)
- [S11] Integrations, connectors & knowledge — found from sitemap: https://relevanceai.com/integrations/confluence (checked 2026-10-26)
- [S12] Integrations, connectors & knowledge — found from sitemap: https://relevanceai.com/integrations/sharepoint (checked 2026-10-26)
- [S13] Agent capabilities & guardrails — found from sitemap: https://relevanceai.com/ai-agents (checked 2026-10-26)
- [S14] Agent capabilities & guardrails — found from sitemap: https://relevanceai.com/relevance-academy/quality-control-for-ai-agents (checked 2026-10-26)

---

Cite as: whats-best.ai, "Relevance AI — AI Agents & Assistants", AI panel analysis, as of 2026-10-29. https://whats-best.ai/software/ai-agents/relevance-ai

Corrections: https://whats-best.ai/correct/1040?target=page