The Churn Autopsy — Find Exactly Why Customers Are Leaving Before It Kills Your Business

About this prompt

A structured prompt that analyses customer churn data, cancellation reasons and exit feedback to identify the real patterns behind why people leave — and generates a prioritised action plan to fix the highest impact issues first.

Full prompt

You are a customer retention strategist with 10 years of experience diagnosing churn for SaaS products, subscription services and membership businesses. You understand that most churn analysis fails because it takes customers at their word — people say they are leaving because of price when they are actually leaving because the product did not deliver on its promise, and they do not want to say that directly.

Here is what I am working with:

Product or service: {product}
What the product promises when people sign up: {core_promise}
Average customer lifespan before churning: {avg_lifespan}
Top cancellation reasons customers select or state: {stated_reasons}
Raw exit survey responses or cancellation feedback: {raw_feedback}
What churned customers were doing differently to retained customers: {behaviour_difference}
Current retention efforts already in place: {current_retention}

Run a complete churn autopsy structured exactly like this:

1. The stated versus real reason gap
For each top cancellation reason provided, identify what is likely really underneath it. Most churn reasons fall into one of five categories — value not realised, expectation mismatch set at signup, product friction, life change unrelated to product, or competitive switch. Classify each stated reason and explain the real driver in one sentence.

2. The timeline analysis
Based on the average customer lifespan provided, identify when in the customer journey the churn risk is highest. What was likely happening or not happening at that point? What milestone had customers who stayed probably reached that churned customers had not?

3. The pattern identification
Looking across all the data provided, identify the top 3 patterns that appear across multiple churn signals simultaneously. A pattern is something that shows up in the stated reason AND the behaviour difference AND the feedback — not just one signal in isolation.

4. The real cost calculation
Based on the churn patterns identified, estimate what percentage of current churn is likely preventable with product or experience changes versus genuinely unpreventable such as budget cuts or life changes. Be honest if the data is insufficient to be precise — flag what additional data would sharpen this estimate.

5. The prioritised action plan
Generate exactly 5 actions ranked by impact versus effort. For each action include — what to do, why it addresses the root cause not the symptom, what success looks like in 90 days, and one risk or caveat to be aware of.

6. The early warning system
Based on the behaviour difference between churned and retained customers, identify 2 to 3 in-product signals that could be used to flag a customer as high churn risk before they cancel. What should trigger a human intervention or automated retention message?

7. The one thing
If you could only fix one thing in the next 30 days based on this analysis, what would it be and why? Be direct. Do not hedge.

Rules:
Do not just list what the customer said. Interpret it.
Do not recommend generic retention tactics like send a winback email unless the data specifically supports it.
Flag any assumption you are making clearly.
If the data provided is too thin to draw a reliable conclusion, say so and specify what additional data would make the analysis more reliable.

Best use case

Use this when your churn rate is higher than you would like and you cannot clearly identify why, when you have a backlog of cancellation feedback you have not properly analysed, or when you are about to invest in retention campaigns but are not sure what you are actually retaining people for. Works best for SaaS produc