Use case
Lead enrichment and routing with AI
Updated September 17, 2026
A sign-up form gives you an email and a company name. This workflow looks the company up, asks Claude how well it fits your ideal customer, and routes the lead based on a score you can read, not a hunch.
How the workflow runs
- The trigger receives the lead.
- An HTTP Request node calls your enrichment API with the company domain.
- A score agent compares the enriched profile with your ideal customer description and returns
scoreandreason. - An If node sends scores of 70 or more to sales and the rest to nurture.
- Merge and Output return the routed lead.
Pass data between nodes
https://api.example.com/companies?domain={{ trigger.email.split('@')[1] }}Ideal customer: B2B software teams of 20 to 500 people.
Company profile:
{{ nodes.enrich.output.body }}
Score the fit from 0 to 100 and explain in one sentence.nodes.score.output.structured.score >= 70Handle a flaky enrichment API
Give the enrich node 3 attempts with backoff, and set its error mode to error_port so leads the API cannot find still reach a manual review branch instead of failing the run. See execution controls.
Keep reading
- If node: branch on a conditionThe If node branches a Novalink workflow on an expression, firing its true or false port so only one path runs and the other is skipped.
- Execution controls: retries, timeouts and error handlingControl how Novalink runs AI workflows: per-node retries with backoff, timeouts and error modes, and run-wide parallelism, timeouts and token caps.
- AI support ticket triageBuild an AI support ticket triage workflow: Claude classifies urgency and topic with structured output, urgent tickets escalate, and the rest get a drafted reply.
Build it on the canvas
Create a free account and run your first workflow in the browser, or run Novalink on your own machine.