Try it now
We'd use it ourselves. It finished the task, "Does it work?" scored 4 or higher, and no criterion scored below 3.
Plain-English guide
Use it, Watch it, Skip it: what our verdicts mean
Every tool BenchSmoke tests ends in one of three verdicts. Here is what each means, the rule behind it, and what to do next.
Updated · By BenchSmoke
The three verdicts
Every tool we test gets exactly one. Here is what each means and what to do with it.
Promising, not ready
It partly finished the task, or has one big weakness such as price, bugs, or a missing feature. Keep an eye on it. We recheck later.
Save your time
It failed the task, wasted time, or is clearly worse than an alternative at the same price.
How we decide
Every tool is tried on a real, small task that fits what it claims to do, in one session. For each tool we score six questions from 1 to 5: does it work, time saved, output quality, ease of use, value, and reliability. Then the verdict rules apply:
| Verdict | What has to be true | What to do |
|---|---|---|
| Use it | It finished the task. "Does it work?" is 4 or higher. No criterion is below 3. | Try it today. |
| Watch it | It partly finished the task, or has one big weakness (price, bugs, a missing feature). | Keep an eye on it. Don't count on it yet. |
| Skip it | It failed the task, wasted time, or is clearly worse than an alternative at the same price. | Save your time. |
- When in doubt, the lower verdict wins. If a tool sits between two verdicts, we pick the lower one and explain why.
- Scores are a guide, not a grade. We use them to reach the verdict. We don't publish them as a rating.
What a verdict is not
- Not a verdict on everything the tool does. It covers the task we tried and the plan we used, and we say what we didn't test, such as team features or a paid tier.
- Not bought. Verdicts are set before anyone checks whether an affiliate program exists. Affiliate links appear only for tools rated Use it, and they are labeled. A sponsor never sees or influences a review.
- Not permanent. A verdict can be revisited. Any change is noted in a later issue, and mistakes are corrected visibly.
- Not given to every tool we mention. Quick Hits are brief notes on launches and updates. They aren't full tests, so they carry no verdict.
The full method, including the standard tasks and our honesty rules, is on How we test. To run the same process on a tool yourself, use the AI coding tool checklist.
Quick answers
What does Use it mean?
We'd use it ourselves. It finished the task, "Does it work?" scored 4 or higher, and no criterion scored below 3.
What does Watch it mean? Is it a bad verdict?
No. Watch it means promising, not ready: the tool partly finished the task, or has one big weakness such as price, bugs, or a missing feature. It is worth rechecking later.
What does Skip it mean?
Save your time. The tool failed the task, wasted time, or is clearly worse than an alternative at the same price.
Can a verdict change?
Yes. A verdict can be revisited, and any change is noted in a later issue.
Do sponsors or affiliate links change a verdict?
No. Verdicts are set before anyone checks whether an affiliate program exists, and a sponsor never sees or influences a review.
More from BenchSmoke
AI coding tool evaluation checklist
14 steps to test an AI coding tool yourself, with a notes template you can copy.
BenchSmoke FAQ
Answers about the newsletter, how we test, the verdicts, and how sponsors and affiliates are handled.
How we test
The full method behind every verdict: the standard tasks, what we record, and our honesty rules.
Get the verdicts every Tuesday and Thursday.
Free, every Tuesday and Thursday, with one clear verdict per tool.