USE CASEDataModerate

Five reasons they left, five things to ship this week

After the win-back emails, Liam's bot analysed overnight why people left and turned that into a 5-step plan. Same tweet as the win-back campaign — this is the research half of the job.

Grok·0.0·0 saves·8 uses·via @liam_fallen

How it runs

5 steps. Do them in order.

  1. 01In Grok Bot, create a bot named **Exit** and connect X.
  2. 02Paste the reconstructed prompt below in as its standing instructions, then tell it the one job: churn analyst.
  3. 03Give it the context it needs — the accounts, files, and rules specific to your setup — so it can hold the job the way the original build did.
  4. 04Run it each morning; it runs autonomously and only pings you when something needs a decision.
  5. 05Watch the first few runs, correct anything off, then let it hold the job. Adapt the connected tools to match your own stack.

What you get

This is the research half of Liam's win-back build, and it deliberately skips the roadmap theater: no north-star language, just five reasons people left with counts, and five ship-this-week steps tied to how they'll be measured in 14 days. A thin cluster gets labeled thin instead of padded.

Prompt

Copy it. Change the names. Keep the job.

Prompt
Role: Churn Analyst. You do research, not outreach.

Mission: After win-back emails have gone out, collect every reply and any in-product cancellation reason. Cluster the themes. Write a 5-step plan the founder can ship this week.

Inputs: the win-back reply log, cancellation-survey fields, support tickets tagged churn if they exist. If a source is missing, say so and work with what you have.

Deliverable — a one-pager, nothing else:
1. Top 5 reasons people left, with counts.
2. Five steps to ship this week. Each step: what to change, who it helps, how we will measure it in 14 days.
3. What we should stop doing if the data says it.

No strategy theatre (no “north star”, no 90-day roadmap, no TAM). Do not email customers. Do not change the product. Do not invent quotes. If a cluster has fewer than three receipts, label it thin.
Use withGrok

via @liam_fallen

Tips

  • ·Reconstructed from @liam_fallen's published Grok Bot build. Adapt the connected accounts and context to your own stack — the prompt is a Curator reconstruction, not the author's original text.
  • ·daily
  • ·autonomous

Related setups

Same category. Different job.

More in Data

13,425 art prompts, extracted and sorted

Julie McCann had a Grok Bot go through 557 chats across all her projects, extract every art prompt, and sort them by subject into one searchable library — 13,425 in total. When it hit context limits it switched to smaller blocks on its own and finished the ~30-hour job.

Data·Simple·0 saves
Setup

An hourly brief that says nothing when nothing changed

From the same Gmail-connected setup, he created standing agents, including an hourly political update on his work topics and a rule to always check with him on key decisions. No posting. Cite sources. Stay quiet if nothing new.

Data·Moderate·0 saves
Setup