Find your leaks: a simple conversion funnel readout
Your store already holds the numbers that show exactly where shoppers fall away; you just line up five stages and read the gaps between them.
Hand your exported analytics to AI and it turns a wall of metrics into one readout that points at the single step worth fixing first.
Fastest path: one prompt, end to end
🤖 AI prompt — paste into ChatGPT / Claude
You are a conversion analyst. Use only the numbers I paste; do not invent any.
Paste from my analytics (last 30 days): sessions [#], product/page views [#], add-to-cart [#], reached checkout [#], completed orders [#].
My history if I have it: the same five numbers from the prior 30 days [paste, or say "none"].
Do this:
1. Build a 5-stage funnel: sessions -> product views -> add-to-cart -> checkout -> order. Show each stage count and the step-to-step conversion % (this stage divided by the previous stage).
2. Flag the biggest single %-drop between two stages; that is where I work first.
3. Compare each step % to MY prior period, not to industry averages. Say which steps got worse and which improved.
4. In one line each, list the usual on-page causes for the leaking step (add-to-cart leak: price, shipping shown late, weak product page; checkout leak: forced account, surprise fees, few payment options).
Do not guess any number. If I gave only one period, say you cannot trend yet and just rank the drops. If any number is missing, ask me for it, never estimate.
Output: the funnel table + the one step to fix first + likely causes.
This note is about reading the funnel. Producing those numbers is a different job: pixel and event setup lives in the conversion-tracking-foundations note under marketing.
Or do it in 4 steps
- Pull the five numbers. From your analytics, export sessions, product views, add-to-cart, reached-checkout, and completed orders for the last 30 days. One row of five numbers is enough to start; you don't need a fancy dashboard.
- Compute step-to-step %, not the overall rate. Each stage divided by the one before it: product views / sessions, add-to-cart / product views, checkout / add-to-cart, order / checkout. The single overall rate hides where the loss happens; the step rates expose it.
- Find the biggest %-drop. The stage-to-stage step with the largest fall is your leak. Fixing a 20% add-to-cart step lifts everything downstream; polishing a step that already converts at 80% barely moves the total.
- Benchmark against yourself, not the industry. Compare this month's step rates to your own prior month. Industry averages are noise: your traffic mix and price point are unique. A step that dropped versus last month is a real regression worth chasing.
Worked example (labeled): 10,000 sessions -> 6,000 product views (60%) -> 1,200 add-to-cart (20%) -> 900 checkout (75%) -> 720 orders (80%).
The add-to-cart step at 20% is the biggest drop by far, so the product page is where you work, not checkout, which already converts at 75%. Last month add-to-cart was 26%, so this is also a real regression, not a baseline. From five numbers you know the one screen to fix, and that it got worse.
Read the steps, not the single number; chase the biggest %-drop and always benchmark against your own history, never an industry average.