AI Blocks › Write emails that get read
Email test planner
Give it your flows and send volumes. Get a ranked test queue with sample sizes and win rules.
Email test planner is a free AI skill for the job "Write emails that get read". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · Made by AI Blocks
What it does
Most email tests teach nothing. A store changes two things, picks a winner on open rate after a few hours and moves on. The email with more opens often earns fewer orders. In one test the source reports, two subject lines had 68 and 69 percent opens but 11.6 and 3.8 percent placed-order rates. This skill plans tests the way a careful buyer would. It writes a hypothesis first and picks tests where your volume is, so a test can finish. It sets the sample size and the winning metric, which is placed orders or revenue and not opens. It sets campaigns to a manual 50/50 split, and tells you when your list is too small to test a given thing. It ranks tests by likely impact so you start with the biggest.
You give it
- Monthly revenue and email list size
- Your flows with recipients per month and revenue per flow
- Campaign send history, with revenue per send if you have it
- Your current popup signup rate (optional)
- Email flow (optional)
You get
- A ranked test queue, each test with a hypothesis
- Sample size and run time for every test
- Set-up notes, the winning metric and the rule for calling a result
- A list of tests to skip for now, and why
Use it when
- You want to test your emails but do not know where to start.
- You ran an A/B test and are not sure the winner is real.
- You have a small list and need to know what you can test and what you cannot.
How to use it
- Start a chat with the skill. Open it if you added it to your AI, or copy it and paste it in.
- Paste your monthly revenue, list size, flows with recipients and your last 8 or more campaigns.
- Pick the first test from the queue and set it up with the notes in the plan.
- Wait for the sample size before you read the result.
- Record each result, then ask the skill to update the queue.