AI Blocks › Test changes to your store
Know which store change to test first, ranked by profit
Give it your store numbers and what shoppers do on key pages. Get a profit baseline and a ranked list of tests.
Find the store change most worth testing is a free AI workflow for the job "Test changes to your store". It guides your AI through several skills in order, passing each step's output to the next.
Free · No account · Made by AI Blocks · Thinking: high
What it does
When a store has traffic but no clear next move, the test list fills up with opinions. Someone wants a new button colour, someone wants a banner. Tests are slow and cost sales while they run, so a wrong pick is expensive. Conversion rate alone also misleads, because cold traffic pulls it down and says nothing about profit. The first step corrects your dashboard numbers for tax, shipping and bots, then works out revenue and gross profit per visitor. That becomes the baseline every idea is sized against. Then it pauses and asks you to watch about 20 recordings each on the product page and the cart, save heatmap screenshots and note any survey answers. The second step turns that into a list of friction, weighted by money. The third splits repairs from tests, writes each test as a hypothesis, drops ideas with thin evidence and picks the one or…
You give it
- A store report with orders, sessions, revenue, tax, shipping and costs for a date range
- After the pause, heatmap and scroll map screenshots, notes from about 20 recordings each on the product page and cart, and any survey answers
- Customer reviews or survey comments (optional)
- Customer words (optional)
You get
- Corrected revenue and gross profit per visitor, a split of cold and warm traffic, and a trust score out of 100 for the numbers
- A friction list ranked by page and money at stake, a test idea for each item and a friction score out of 100
- A list of repairs to make now that need no test
- A leaderboard of test ideas, the one or two to run first and the reasons the rest wait
Use it when
- You have traffic but no clear idea what to test on your store next.
- Your team keeps suggesting changes and you need a fair way to choose.
- Your conversion rate moved and you want to know whether it affected profit.
Steps
- Store profit per visitor calculator: Paste your Shopify and analytics numbers. Get corrected revenue and gross profit per visitor, and a trust score.
- Heatmap and recording reader: Paste your heatmap and recording notes. Get the friction found, a test idea for each, and a friction score.
- Test idea ranker: Paste your research findings. Get each test written as a hypothesis and ranked by evidence and profit.
How to use it
- Download the handoff pack. It's a zip with HANDOFF.md and every skill file this workflow uses.
- Give the folder to Claude (in Claude Code, or as files in a Claude Project) and say "Follow HANDOFF.md".
- Give it a store report for a recent range, with sessions, orders, tax and shipping.
- When it pauses, watch about 20 recordings each on the product page and cart. Save heatmap screenshots and write down what you saw.
- Make the repairs list first. Run only the tests it names, one at a time.