How to integrate Unify with Relevance AI
To integrate Unify with Relevance AI, send a POST request from a Unify Play to the tool’s configured webhook trigger and map the received fields into the intended action. Use this connection to start a Relevance AI agent with a selected prospect record.[1][2][3]
- Connection
- Via webhook
- Automation
- Tool
- Relevance AI
- Relevance AI
- Data flow
- Unify → Relevance AI
- JSON payload
At a glance
- Via webhook: Relevance AI receives the handoff
- POST request with an explicit JSON contract and destination authentication
- Test scenario: Keep conversations separate
When this workflow makes sense
Start with the decision your team needs to make, then choose the connection that supports it.
An agent is useful when the next action requires interpreting context, but that interpretation needs boundaries. State what it may research, what it may save, and what still requires human review. Start with a factual handoff brief before allowing wider actions.
A good fit when
Teams that want to start a Relevance AI agent with a selected prospect record and have an owner for the receiving process. The destination account must permit the documented request and the chosen downstream action.
Reconsider the plan when
The receiving team has not agreed how this Relevance AI operation should behave when the same record arrives again. Resolve the matching rule before enabling retries.
Design the handoff
Use this framework to agree on the input, the next action, and who owns the result.
Suggested workflow design, based on the documented capabilities below.
| Stage | Decision to make |
|---|---|
| Business decision | Start a Relevance AI agent with a selected prospect record |
| Receiving contract | Agree which of the 5 example data fields the receiving process actually needs. Set missing-value and identity rules in the field mapping below. |
| Execution boundary | Give one owner access to inspect the Relevance AI request and its resulting action. |
| Completion | Decide who can confirm the Relevance AI outcome and recover a failed or repeated handoff. Use the completion check below as the acceptance criterion. |
Explore the scenarios
Choose the situation that fits your team and work through the proposed plan.
Open a scenario to explore its plan. Checklists track your selections in this page only.
Start a Relevance AI agent with a selected prospect record
Start with one controlled record and one intended Relevance AI outcome. Keep the test destination separate from live operations.
The receiving owner must distinguish accepted input from completed work. Agree on the evidence required before treating this handoff as successful.
- Prepare the destination using the Relevance AI setup checklist
- Resolve the sample fields for one known record and inspect the request before sending it
- Trace that handoff through the receiving operation and compare the actual destination values with the source
Keep conversations separate
Keep conversations separate: submit two prospect records and inspect their Thread ID mappings. Confirm that one account’s context cannot be reused accidentally in the other account’s task.
Decide whether this case should be held for review, corrected, or retried. Make that decision before running Relevance AI against live records.
- Isolate the test in a controlled destination and record the starting state
- Run the edge case described above and capture the actual response and destination state
- Recover the record once, then check that recovery did not create unintended additional work
Illustrative workflow, not a preconfigured automation. Complete these checks in your own account.
Define success before launch
Inspect the data and the intended outcome separately. Connection status is only the first check.
Expected, not tested
Start a Relevance AI agent with a selected prospect record
Inspect the agent task input, any tool calls, and the final artifact. Confirm that a repeated handoff follows the configured Unique ID behavior.
Documentation-based reference. This workflow has not been tested end to end. Use a record or event you control before enabling live activity.
Questions answered
Common decisions when connecting Unify and Relevance AI.
What is the difference between Thread ID and Unique ID?
Thread ID controls conversation grouping. Unique ID helps prevent duplicate processing of an incoming event. Choose each mapping for its own purpose.[3]
Limits & decisions
Check these boundaries before designing a live workflow around the connection.
What the connection supports
Understand the connection, the data involved, and the behavior that matters for a Unify + Relevance AI workflow.
POST / JSON
| Decision | What to know |
|---|---|
| Unify action | Outbound HTTP request with configurable method, authentication, headers, and body[1] |
| Receiving operation | Add a Webhook trigger to the agent and copy the generated URL. Set a message template that references the incoming payload.[3] |
| Response handling | Webhook response bodies are not available as inputs to later Unify Play actions[2] |
Technical setup reference
References and requirements for the Relevance AI connection.
Connection requirements and setup checklist
Before connecting
- A Unify Play with a Webhook action and a controlled record for testing
- Access to the intended Relevance AI environment and the permissions required by the cited API or trigger
- The destination identifiers, credentials, and schema used by the request below
- An agreed record identity, duplicate policy, and owner for failed handoffs
Prepare the Relevance AI destination
Add a Webhook trigger to the agent and copy the generated URL. Set a message template that references the incoming payload.[3]
Configure the destination action
Map the trigger Unique ID from handoff_id and choose Thread ID only when requests should share a conversation. Give the agent an explicit output destination and review boundary.[3]
Configure the Unify request
Use POST, the endpoint and authentication below, and JSON body format in the Webhook action. Replace setup placeholders, then inspect Expand preview with a controlled record. The receiver must respond within 8 seconds; webhook response bodies are not exposed to later Play steps.[1][2]
Validate the complete handoff
Run the success and edge-case scenarios in Explore the scenarios with a controlled record. Save the request identity and destination evidence, then apply the completion check before enabling live traffic.[3]
Request configuration
- Method
POST- URL pattern
https://YOUR_RELEVANCE_AI_WEBHOOK_URL- Headers
Content-Type: application/json- Authentication
- Use the URL generated for your agent trigger and keep it private. Configure any access requirements shown by the trigger; this guide does not assume that an arbitrary Bearer header is validated.
Choose JSON as the body format. Unify sets Content-Type automatically. The receiver must respond within Unify’s 8-second timeout; request responses are not passed to later Play steps.[2]
Replace URL placeholders with the actual receiver URL from your account or deployed application. These example URLs are not working endpoints.
Example handoff payload
Start with a fictional prospect. Validate the resolved values before using a real record.
Inspect the sample payload and field mapping
{
"event": "prospect_handoff",
"handoff_id": "example-handoff-001",
"prospect_name": "Alex Morgan",
"prospect_email": "alex@example.com",
"company_name": "Example Company",
"company_domain": "example.com"
}Fictional data. Replace sample values before using a real Play.
Illustrative JSON for the documented receiving operation. All prospect values are fictional. Replace credentials, identifiers, schema-specific fields, and fixed example values before a controlled test. Use the field mapping to distinguish record data from fixed settings. Select available fields in your Play and inspect Expand preview; do not replace configuration values with prospect data.[1][2][3]
| Payload field | Value to send |
|---|---|
event | Keep the literal "prospect_handoff" required by this example request; change it only when deliberately changing the receiving contract. If missing: Keep this configured value in the request; do not replace it with a blank prospect field. |
handoff_id | Supply a stable identifier for this logical handoff. Reuse it on a retry, and use a different identifier for a genuinely new event. Do not assume Unify automatically creates this sample field. If missing: Hold the handoff until the receiver or upstream workflow can establish its identity. |
prospect_name | Select the available full name field for the controlled person or company record in Unify. Keep the receiving field a string. If missing: Skip or hold the handoff if this value is required by the destination; omit it only when the destination contract permits omission. Do not send a guessed value. |
prospect_email | Select the available email field for the controlled person or company record in Unify. Keep the receiving field a string. If missing: Skip or hold the handoff if this value is required by the destination; omit it only when the destination contract permits omission. Do not send a guessed value. |
company_name | Select the available company name field for the controlled person or company record in Unify. Keep the receiving field a string. If missing: Skip or hold the handoff if this value is required by the destination; omit it only when the destination contract permits omission. Do not send a guessed value. |
company_domain | Select the available company domain field for the controlled person or company record in Unify. Keep the receiving field a string. If missing: Skip or hold the handoff if this value is required by the destination; omit it only when the destination contract permits omission. Do not send a guessed value. |
Keep event identity and conversation identity separate. Use the prospect fields as supplied context and map handoff_id to the Unique ID setting when deduplicating.
Troubleshooting
Start with the stage where the expected data or action stops appearing.
The request is rejected
Use the URL generated for your agent trigger and keep it private. Configure any access requirements shown by the trigger; this guide does not assume that an arbitrary Bearer header is validated. Check the receiving account, permissions, endpoint, and required fields against the cited documentation.[3]
The values or destination are wrong
Compare the resolved JSON with the field-by-field contract in Example handoff payload. Check identifiers against the selected Relevance AI environment, and distinguish an absent source value from an intentional static value. Fix the mapping before repeating the write.[3]
Receipt does not match the expected result
Find the request in the receiving logs, then inspect the actual Relevance AI resource or execution. If the request was accepted but no result exists, investigate the destination failure before retrying. Keep the same logical handoff identity when recovering a repeat.[3]
Sources & review
Official documentation reviewed on September 15, 2026.
Technical claims link to primary documentation from Unify and the destination tool. Scenarios, field choices, and recovery plans are editorial implementation examples. Documentation review is separate from execution testing.
Documentation reviewed; end-to-end execution not tested. No named technical reviewer has signed off on these articles yet.
Content updated: . Report a correction to Unify with the tool name, source URL, and behavior you observed.
Keep exploring
