← All runbooks
gooseworks-ai / composites-expansion-signal-spotter

Expansion Signal Spotter

Monitor existing customer accounts for expansion signals and turn those signals into a prioriti

agent codexmodel gpt-5.5snapshot python312-uveval programmatic8 stepsv1.0.0

Deploy Expansion Signal Spotter 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.

Deploy on jetty.io →View source
The shape of the run

8 steps · start to finish.

  1. 1
    Step 1

    Environment Setup

    ▶
    1. Create /app/results and confirm it is writable.
    2. Resolve all required inputs: customer list, product tiers, and any account filters.
    3. Normalize account records so each account has a company name, domain, current plan, current revenue, seat or usage information when available, and primary contact.
    4. Write a setup entry into /app/results/validation_report.json if any required input is missing, then stop before scanning.
  2. 2
    Step 2

    Load Account Data

    ▶

    Read the customer list from the supplied CSV, sheet, CRM export, or structured text. For each account, capture company name, domain, primary contact LinkedIn URL, current plan or tier, MRR or ARR, seats, usage, and any notes that affect expansion eligibility.

  3. 3
    Step 3

    Detect Expansion Signals

    ▶

    For each eligible account, run targeted searches for recent hiring, funding, revenue, product launch, partnership, geographic expansion, and category-specific news signals. Include searches such as:

  4. 4
    Step 4

    Score Opportunities

    ▶

    Compute an expansion score for each account:

  5. 5
    Step 5

    Generate Talk Tracks

    ▶

    For every Hot and Warm opportunity, write a talk track that connects the detected signal to the customer’s business goal, identifies the expansion type, estimates additional MRR when possible, names the timing reason, and lists the main risk or blocker.

  6. 6
    Step 6

    Produce Required Outputs

    ▶

    Write `/app/results/expansion_signal_report.md` with the weekly report, including accounts scanned, total expansion pipeline identified, Hot/Warm/Watch summaries, account-level recommendations, trends, and the top three priorities for the week.

  7. 7
    Step 7

    Validate Outputs

    ▶

    Run structural checks before finishing:

  8. 8
    Step 8

    Iterate on Errors (max 3 rounds)

    ▶

    If any required output is missing, any JSON file fails to parse, or a Hot/Warm opportunity lacks a signal source, perform up to max 3 rounds of targeted repair. Re-run only the failed scan, scoring, or output-generation step, then repeat validation. After 3 rounds, write remainin