Adoption Decline Diagnostic
When the churn alarm fires, this workflow tells you which lever is broken — not just that something is.
Sample output for this workflow will appear here once it is captured.
Run the workflow in Claude, ChatGPT, or HG Insights Playground using the buttons below to see real output.
Overview
When the churn-prediction screen flags a Yellow or Red on a customer, you need to know why. This workflow runs the adoption-decline forensics: which products dropped, in which geos and functions, was the customer hit by layoffs or a restructuring, did a competitor land. The output is a one-page diagnostic with a named root-cause hypothesis, not a list of correlated signals.
Use cases
From 'something's wrong' to 'here's what broke'
Your CSM gets a Yellow on an account and forwards it to the AE with 'churn risk, let's discuss'. This workflow turns that into 'EMEA office stopped using us two months ago, right after the new RVP started — competitor X landed in the same window'. The conversation goes from vague worry to specific action.
Function-level decline is a different play than broad decline
If only Engineering lost intensity but Sales held steady, that's a champion / team-turnover story (engage the new VP). If every function dropped, that's a financial-stress or strategic-substitution story (escalate to exec sponsor). FAI makes that distinction visible.
View workflow prompt
# Adoption Decline Diagnostic
## Parameters
- `{{domain}}` *(required)* — Customer company domain. Example: `acme.com`
- `{{your_product}}` *(required)* — Your product or category. Example: `HG Insights`
## Purpose
A churn-prediction workflow flagged {{domain}} as Yellow or Red. This workflow diagnoses *why* adoption is declining for {{your_product}} — function-level, geo-level, competitive, or financial — and proposes a single named root-cause hypothesis with evidence, not a correlated-signals list.
## Process
1. **Decline shape** — `company_install_time_series` for {{domain}} filtered to {{your_product}} (or its category via `get_product_category`), `timeRange: last_12_months`. Look at `data_points[]`: is the decline a steady drift or a cliff? Identify the inflection month.
2. **Function-level slice** — `company_fai` with {{your_product}} as the product. Capture current `departmentUsageShare` per function (Engineering / Sales / IT / etc.). The tool does NOT expose a temporal trend — to infer "did a function drop", compare today's FAI distribution against your CRM / prior brief baseline if you have one; otherwise treat as current-state context, not a delta signal.
3. **Competitive substitution** — `company_technographic` filtered to {{your_product}}'s category. Any vendor with `firstVerifiedDate` just before the inflection month? Cite `productLocations` to gauge install breadth.
4. **Financial / org stress** — `sec_filing_section` 10-K Item 1A (Risk Factors) and Item 7A. Layoffs ≥5%, restructuring, guidance miss, going-concern. Also `web_search` for "{{domain}} layoffs" or "{{domain}} restructuring" if non-public.
5. **Hypothesis** — pick ONE root cause from {competitor substitution, function-team turnover, broader RIF, restructuring/M&A, product fit erosion}. State it in one sentence; cite the 2-3 strongest signals that support it.
## Output Format
- `# 🔎 {{domain}} — Adoption Decline Diagnostic`
- `**Inflection month**: <month> — <steady drift|cliff>`
- `## Root-cause hypothesis` — one sentence, bolded
- `## Supporting evidence` — 2-3 bullets, each cites its tool
- `## Recommended next step` — one action tied to the hypothesis (e.g., "Engage the new VP Engineering who joined 3 months ago — the decline is concentrated in their function")
## Quality Checklist
- One named hypothesis, not a menu
- Inflection month identified from `data_points`, not guessed
- Competitor substitution cites `firstVerifiedDate`, not just product name
- Cite every signal at its source boundary
- Cap tool calls at ~12