Score Leads on Firmographic Signals with the Clay CLI

Compute a composite lead score from real company data — headcount, follower growth, business stage — using the clay CLI to run Clay's Enrich Company routine, then scoring the result yourself.

What you will build

A short bash + jq workflow: run Enrich Company via the CLI, extract real firmographic signals, and compute a weighted composite score locally.

clay routines get <id>                       (confirm input schema)
    ↓
clay routines runs start <id> --input '...'  (submit company)
    ↓
clay routines runs get <run-id> --wait 60    (poll until complete)
    ↓
jq-based scoring script                       (weights applied locally)

AI Prompt

Using the clay CLI, compute a composite lead score for a company using
Clay's "Enrich Company" managed function for firmographic signals.

Requirements:
- Find the real routine id for "Enrich Company" via `clay routines list`
  (source: managed).
- Confirm the real input schema with `clay routines get <id>` before
  submitting -- the required input key is "Company Identifier" (accepts a
  domain or LinkedIn URL).
- The result is keyed "Enrich Company" (matches the display name, unlike
  some other Clay routines). It contains employee_count, follower_count,
  total_funding_amount_range_usd, and derived_datapoints.business_stage --
  it does NOT contain headcount_growth_pct or ai_fit_score. Do not use
  those invented field names.
- Map: firmographic <- employee_count (direct); momentum <- follower_count
  (a proxy for growth, not a native rate -- say so); stage <-
  derived_datapoints.business_stage via your own lookup table.
- Normalize each signal to 0-100, apply your own weights, and compute the
  composite score in your script -- this scoring policy lives in your code,
  not in Clay.
- Run the verification step below before finishing.

Using the clay CLI, compute a composite lead score for a company using
Clay's "Enrich Company" managed function for firmographic signals.

Requirements:
- Find the real routine id for "Enrich Company" via `clay routines list`
  (source: managed).
- Confirm the real input schema with `clay routines get <id>` before
  submitting -- the required input key is "Company Identifier" (accepts a
  domain or LinkedIn URL).
- The result is keyed "Enrich Company" (matches the display name, unlike
  some other Clay routines). It contains employee_count, follower_count,
  total_funding_amount_range_usd, and derived_datapoints.business_stage --
  it does NOT contain headcount_growth_pct or ai_fit_score. Do not use
  those invented field names.
- Map: firmographic <- employee_count (direct); momentum <- follower_count
  (a proxy for growth, not a native rate -- say so); stage <-
  derived_datapoints.business_stage via your own lookup table.
- Normalize each signal to 0-100, apply your own weights, and compute the
  composite score in your script -- this scoring policy lives in your code,
  not in Clay.
- Run the verification step below before finishing.

Prerequisites

The clay CLI on PATH, authenticated
jq

1. Find and confirm the routine

clay routines list

clay routines list

Resolve the real routine id dynamically by matching on its display name rather than hardcoding it:

ROUTINE_ID=$(clay routines list | jq -r '.data[] | select(.name=="Enrich Company") | .id')
clay routines get "$ROUTINE_ID"

ROUTINE_ID=$(clay routines list | jq -r '.data[] | select(.name=="Enrich Company") | .id')
clay routines get "$ROUTINE_ID"

Real confirmed input schema requires "Company Identifier" (a domain or LinkedIn URL).

2. Run it

clay routines runs start "$ROUTINE_ID" \
  --input '{"items":[{"id":"clay-1","inputs":{"Company Identifier":"clay.com"}}]}'

clay routines runs start "$ROUTINE_ID" \
  --input '{"items":[{"id":"clay-1","inputs":{"Company Identifier":"clay.com"}}]}'

clay routines runs get <run-id> --wait 60

clay routines runs get <run-id> --wait 60

3. Score the result locally

#!/bin/bash
STAGE_SCORES='{"Seed":20,"Early Stage":40,"Growth Stage":80,"Mature":60,"Public":50}'

RESULT=$(clay routines runs get "$1" --wait 60 | jq '.data[0].result["Enrich Company"]')

EMPLOYEES=$(echo "$RESULT" | jq -r '.employee_count // 0')
FOLLOWERS=$(echo "$RESULT" | jq -r '.follower_count // 0')
STAGE=$(echo "$RESULT" | jq -r '.derived_datapoints.business_stage // "Unknown"')
STAGE_SCORE=$(echo "$STAGE_SCORES" | jq --arg s "$STAGE" '.[$s] // 30')

FIRMO_NORM=$(python3 -c "print(min($EMPLOYEES/100, 100))")
MOMENTUM_NORM=$(python3 -c "print(min($FOLLOWERS/2000, 100))")

COMPOSITE=$(python3 -c "print(round($FIRMO_NORM*0.4 + $MOMENTUM_NORM*0.3 + $STAGE_SCORE*0.3, 2))")

jq -n --arg domain "clay.com" --argjson score "$COMPOSITE" \
  --argjson employees "$EMPLOYEES" --argjson followers "$FOLLOWERS" \
  '{domain: $domain, composite_score: $score, employee_count: $employees, follower_count: $followers}'

#!/bin/bash
STAGE_SCORES='{"Seed":20,"Early Stage":40,"Growth Stage":80,"Mature":60,"Public":50}'

RESULT=$(clay routines runs get "$1" --wait 60 | jq '.data[0].result["Enrich Company"]')

EMPLOYEES=$(echo "$RESULT" | jq -r '.employee_count // 0')
FOLLOWERS=$(echo "$RESULT" | jq -r '.follower_count // 0')
STAGE=$(echo "$RESULT" | jq -r '.derived_datapoints.business_stage // "Unknown"')
STAGE_SCORE=$(echo "$STAGE_SCORES" | jq --arg s "$STAGE" '.[$s] // 30')

FIRMO_NORM=$(python3 -c "print(min($EMPLOYEES/100, 100))")
MOMENTUM_NORM=$(python3 -c "print(min($FOLLOWERS/2000, 100))")

COMPOSITE=$(python3 -c "print(round($FIRMO_NORM*0.4 + $MOMENTUM_NORM*0.3 + $STAGE_SCORE*0.3, 2))")

jq -n --arg domain "clay.com" --argjson score "$COMPOSITE" \
  --argjson employees "$EMPLOYEES" --argjson followers "$FOLLOWERS" \
  '{domain: $domain, composite_score: $score, employee_count: $employees, follower_count: $followers}'

4. Verify the result

This example was tested live against clay.com.

clay routines runs start "$ROUTINE_ID" \
  --input '{"items":[{"id":"clay-1","inputs":{"Company Identifier":"clay.com"}}]}'

clay routines runs start "$ROUTINE_ID" \
  --input '{"items":[{"id":"clay-1","inputs":{"Company Identifier":"clay.com"}}]}'

{ "routineRunId": "run_0tjzn5ngCGTiFr87URj", "mode": "inline", "status": "in_progress" }

{ "routineRunId": "run_0tjzn5ngCGTiFr87URj", "mode": "inline", "status": "in_progress" }

clay routines runs get run_0tjzn5ngCGTiFr87URj --wait 60

clay routines runs get run_0tjzn5ngCGTiFr87URj --wait 60

Real confirmed fields in the result:

{
  "employee_count": 1490,
  "follower_count": 182849,
  "derived_datapoints": {"business_stage": "Growth Stage"},
  "total_funding_amount_range_usd": "Funding unknown"
}

{
  "employee_count": 1490,
  "follower_count": 182849,
  "derived_datapoints": {"business_stage": "Growth Stage"},
  "total_funding_amount_range_usd": "Funding unknown"
}

Applying the scoring formula above: firmographic_norm = min(1490/100, 100) = 100, momentum_norm = min(182849/2000, 100) = 100, stage_score = 80 (Growth Stage) → composite_score = 100×0.4 + 100×0.3 + 80×0.3 = 94.0.

How it works

Enrich Company returns real firmographic data in one call; the composite score itself is computed locally with jq/python3, not by Clay. Keeping the scoring formula in your own script (not in the routine) means it's reviewable and changeable without touching Clay's configuration.

Common issues

`jq: error: null (null) has no keys`

Cause: reading derived_datapoints.business_stage on a company where that field is genuinely absent.

Fix: use // "Unknown" fallbacks throughout, and default the stage score to a neutral value (e.g. 30) rather than failing.

Composite score assumes `headcount_growth_pct` or `ai_fit_score` exist

Cause: assuming Clay's Enrich Company output includes pre-built growth-rate or AI-fit fields. It does not.

Fix: derive momentum from follower_count (a real, present field) and be explicit in your script comments that it's a proxy, not a true growth-rate metric.

Next steps

Source the account list first — see the Clay TAM-by-industry CLI example
Feed hiring-momentum signals into the score — see the Clay hiring-signals outbound CLI example
Deep-dive research on high scorers — see the Clay company-news research CLI example

Verification

verification:
  status: verified
  tested_at: "2026-08-18"
  cli_version: "0.7.0"
  auth_method: "clay login --device"
  command: "clay routines runs start function:t_0tjqm9uSZcH4rf4MDrK --input '{\"items\":[{\"id\":\"clay-1\",\"inputs\":{\"Company Identifier
  expected_result: "employee_count=1490, follower_count=182849, business_stage='Growth Stage'; computed composite_score=94.0 with the documented formula"

verification:
  status: verified
  tested_at: "2026-08-18"
  cli_version: "0.7.0"
  auth_method: "clay login --device"
  command: "clay routines runs start function:t_0tjqm9uSZcH4rf4MDrK --input '{\"items\":[{\"id\":\"clay-1\",\"inputs\":{\"Company Identifier\":\"clay.com\"}}]}' && clay routines runs get run_0tjzn5ngCGTiFr87URj --wait 60"
  expected_result: "employee_count=1490, follower_count=182849, business_stage='Growth Stage'; computed composite_score=94.0 with the documented formula"