How we score
Every tool on this site gets one number out of 100 and one of four verdicts. Same criteria, same weights, every time. This page is the whole rubric, so you can disagree with a score and know exactly where.
The five criteria
Output quality (30 points). How usable the result is without heavy editing, how consistent it is, and how it compares with the best tools for the same job. This is the biggest weight because a tool that produces mediocre work is expensive at any price.
Time to value (20 points). How fast a non-technical person gets something genuinely useful after signing up. Sign-up friction, onboarding, templates and defaults, learning curve. Minutes score high. “Book a demo” scores low.
Price fairness (20 points). Not “is it cheap” but “is it fair for what a freelancer or a small team actually gets”. Free-plan generosity, the entry paid plan, seat pricing, overage fees, and whether the advertised price only exists on annual billing.
Reliability and trust (15 points). What the vendor documents about your data (training on customer content, retention, certifications), uptime history, company stability, support response, refund policy.
Fit and integrations (15 points). Whether it plugs into the apps you already use: Google Workspace, Slack, Notion, Zapier, WordPress, Shopify and the like. Import, export, browser extension, mobile app, API for the technical folks.
The four verdicts
- Top pick (80 to 100). Buy it if it matches your use case. We would pay for it ourselves.
- Solid (70 to 79). Good value with a few caveats. Read the “Skip it if” column before you pay.
- Situational (65 to 69). Worth it only for a specific need or budget. The three scenarios on the scorecard tell you whether you are that case.
- Skip (under 65). Better options exist at the same price. We say which.
One hard rule on top of the maths. If reliability and trust scores 40 or lower (a documented pattern of charges people could not stop, unclear refunds, or terms that put your content at risk) the final score is capped at 69, so the tool can never be Solid or a Top pick however good its output. A tool that is hard to stop paying for is not a recommendation on a site whose whole point is deciding before you pay. The scorecard shows when the cap applies.
What we check and what we don’t
Prices, plan limits and free-plan caps are taken from the vendor’s own pricing page and from public directories, and rechecked when we learn a price has moved. Prices change. Check the vendor before you buy.
Scores are editorial opinions built from documented facts. We do not run controlled benchmarks and we do not pretend to. What you get is a buying decision, written down, with its reasons.
How the site makes money
Some outbound links are affiliate links. If you buy through one, the vendor pays us a commission at no extra cost to you. Affiliate status never changes a score, a verdict or a position in a list. Tools with no affiliate program are scored with exactly the same rubric, and several of our top picks pay us nothing.
How AI is used
Our reviews are written by editors with AI assistance. The raw material is documented facts: the vendor’s own pricing and documentation pages, and other public sources. An AI model drafts the review into our structure and proposes the five criteria scores against fixed, written briefs. An editor checks the claims against the source, can adjust a score with a reason shown on the scorecard, and decides what gets published. Nothing goes live without a person behind it. When we get something wrong, that is on us, not on the model. Tell us and we will fix it.