Method
How we test
Every verdict on this site comes from the same test and the same rules. They're published here so you can check our work, or run the numbers for your own business.
The test
Each tool runs for 30 days inside a real workflow a small business actually does. Every run is logged with its date and result. The workflow is published with the verdict so you can rebuild it.
Five measured inputs
| Input | How it's measured | Why it matters |
|---|---|---|
| Time saved per week | Minutes the task took before vs. after, logged per run | The thing you're actually buying |
| Setup cost | Hours from signup to the first fully working run | The cost most reviews skip |
| Reliability | Successful runs divided by total runs over 30 days | An automation that fails quietly costs more than none |
| Monthly cost | Price of the plan the workflow actually needed | Free tiers rarely survive real use |
| Exit cost | Can your workflows and data be exported? Yes, partly or no | Lock-in is a cost you pay later |
The one number
Cost per hour saved = monthly cost ÷ (hours saved per week × 4.33)
A $29-a-month tool that saves 2 hours a week costs about $3.35 per hour saved. Compare that with what an hour of your time is worth.
The rules
Cost per hour saved under $25, reliability at least 95%, and setup under 4 hours.
Passes two of the three. Re-tested after its next major release.
Passes one or none. Published anyway, with the reason.
$25 an hour is our default. Your hour may be worth more or less, so the calculator below lets you change it.
Run the numbers for your business
Move the sliders to see the verdict a tool would get with your numbers.
What we don't do
- We don't accept payment for verdicts or placement.
- We don't rank tools by commission. Tools without an affiliate program get tested the same way.
- We don't publish a verdict without the 30-day log behind it.
Every verdict is dated. When a tool ships a major release, it's re-tested and the date updates.