Editorial standards
How we review tools
Every rating on this site comes from the same process: build something real with the tool, then evaluate it against a fixed set of criteria. Here's exactly what that looks like.
The core question we're answering
This site evaluates tools through one lens: how useful is this for a non-technical founder or product manager? Not raw engineering power. Not developer experience. Not how impressive the demo is.
That means our ratings will differ meaningfully from developer-focused review sites. A tool that requires terminal access to deploy, or that produces code requiring cleanup by a developer before it runs, gets a lower non-coder rating — even if engineers love it.
The non-coder rating (●●●●○)
Every tool gets a 1–5 dot rating. Here's what each level means:
How a tool gets reviewed
Reviews follow a standard sequence:
1. Build a reference project
Every tool is evaluated by building a real project with it — typically a basic SaaS with user authentication, a dashboard, and data persistence. This is the same spec used across all build-tier tools so ratings are comparable. For run-tier tools (marketing, automation, analytics), we build or configure a real workflow, not a demo.
2. Evaluate against fixed criteria
After building, the tool is scored across six dimensions:
- Time to first working output — How long from signup to something real running?
- Abstraction quality — How much complexity is hidden vs. exposed?
- Failure recovery — When things break (they always do), how hard is it to fix?
- Deployment story — Can a non-coder get it live without help?
- Pricing transparency — Is it predictable, or do costs spike unexpectedly?
- Output quality — Does the generated code/content actually work without cleanup?
3. Retest after major updates
Tools are re-evaluated when they ship significant changes. Ratings change — up or down — when the product changes. The last_updated date on each tool page reflects when we last tested it, not when we published the original review.
What we don't do
Sponsored placements. Tools don't pay to appear on this site. We've been approached by vendors offering "featured listing" arrangements and declined all of them.
Press release reviews. Vendor-provided demo environments and curated use cases are excluded from ratings. We test the actual product the way a real user would encounter it.
Rounded scores. If a tool is 3 out of 5, the page says 3. Not 4 because the vendor has an affiliate program and not 2 because a competitor paid for advertising.
Manufactured consensus. Comparisons have explicit winners. "It depends" is sometimes the honest answer, but that's stated directly — not used as a hedge to avoid taking a position.
Affiliate links
Some tool pages include affiliate links. When you sign up through one, we earn a commission at no cost to you. Affiliate status is marked on every page where it applies.
Affiliate arrangements are accepted only for tools we'd recommend anyway. A tool rated 3/5 stays 3/5. Accepting an affiliate deal doesn't change our rating, and declining one doesn't either. We've turned down programs for tools we don't believe in.
How tools are selected for review
We prioritize tools that:
- Are being actively used by founders in our audience
- Have shipped meaningful updates that change our prior assessment
- Fill a gap in our current coverage
- Are being heavily promoted in the founder community and warrant a clear-eyed take
We don't review tools we can't test thoroughly. If something shows up on the site, it's been used — not just demoed.
Corrections and updates
If something is wrong, reply to the newsletter or post in the comments section on the relevant page. We update ratings and correct errors publicly. Prior versions aren't hidden — the updatedDate in the page header tells you when the content changed.