← All runbooks
gooseworks-ai / composites-inbound-lead-enrichment

Inbound Lead Enrichment

This runbook enriches incomplete inbound lead records into qualification-ready lead profiles. It researches each company, identifies the lead role and seniority, discovers relevant stakeholders, checks CRM or pipeline relationships, and emits structured outputs for outreach or ro

agent codexmodel gpt-5.5snapshot python312-uveval programmatic11 stepsv1.0.0

Deploy Inbound Lead Enrichment to your jetty.io

One-click installs this runbook into a collection on your Jetty account. You can run it from the Spot dashboard, schedule it, or pipe inputs in via the API.

The shape of the run

11 steps · start to finish.

  1. 1
    Step 1

    Environment Setup

    1. Create results_dir if it does not exist.
    2. Confirm leads_input is present and readable.
    3. Confirm which enrichment tools and CRM sources are available for the run.
    4. Initialize validation_report.json with setup, input, enrichment, output, and final verification stages.
    mkdir -p "${results_dir:-/app/results}"
    test -n "${leads_input:-}" || echo "leads_input must be supplied by the orchestrator"
    
  2. 2
    Step 2

    Configure Client Context

    Establish enrichment preferences before processing the first lead. On subsequent runs for the same client, load the saved configuration silently when available.

  3. 3
    Step 3

    Assess Data Gaps

    For each lead, inventory known fields and classify missing data. Track whether each record needs company research, person research, stakeholder discovery, CRM relationship checks, or human review. Emit a gap inventory containing known fields, missing fields, enrichment depth, and

  4. 4
    Step 4

    Company Research

    Research the organization connected to each lead. Capture company name, website, industry, size, location, growth signals, business model, recent news, and credible evidence URLs or notes. For personal email domains, infer the company only from explicit evidence, such as the lead

  5. 5
    Step 5

    Person Research

    Identify the person's role, seniority, function, likely buying influence, location, and profile evidence. Resolve aliases carefully and avoid merging records unless there is high-confidence evidence that the identity matches the inbound lead.

  6. 6
    Step 6

    Stakeholder Discovery

    Find relevant stakeholders at the same company based on buyer personas, deal context, and likely buying committee structure. Prioritize stakeholders by function, seniority, public relevance to the problem, and relationship proximity.

  7. 7
    Step 7

    Relationship Check

    Check the CRM, pipeline export, or relationship source for existing contacts, accounts, opportunities, open deals, past meetings, owner assignments, and known relationships. Do not overwrite CRM fields directly unless the orchestrator has explicitly granted update permission.

  8. 8
    Step 8

    Compile Outputs

    Produce `enriched_leads.json`, `enriched_leads.csv`, and `enrichment_report.md`. Each enriched lead record should include:

  9. 9
    Step 9

    Iterate on Errors (max 3 rounds)

    If validation fails or any required output is missing, perform up to max 3 rounds of targeted repair:

  10. 10
    Step 10

    Write Summary

    Write `summary.md` with run date, input count, enrichment depth, available tools, CRM source, output paths, records requiring human review, and any unresolved data-quality issues.

  11. 11
    Step 11

    Final Checklist

    Before completing the run, verify every required output file exists and is non-empty, and validate that JSON and CSV outputs parse.