The Short Version
Public pages use five simple labels: Best Match, User Love, Momentum, Value, and Evidence Level. The internal blended Master Rating stays out of the public UI because it is less useful than showing what changed and which evidence supports the recommendation.
The first answer to: should this tool be on your shortlist for this use case?
Use-case fit, workflow fit, deployment needs, pricing, constraints, and the available evidence.Whether the market signal feels positive, credible, and durable.
Review signals, source quality, sentiment summaries, community discussion, and reputation evidence.Whether the tool is currently moving, improving, or getting fresh attention.
Launches, changelogs, GitHub activity, Product Hunt activity, creator coverage, and fresh news.Whether the pricing and limits make sense for the job.
Free tier, paid plan clarity, local/open-source options, useful limits, and feature depth.How much source support sits behind the recommendation.
Signal count, source diversity, freshness, and whether the source is official, verified, or third-party.Where The Receipts Come From
We combine catalog research with source signals such as official changelogs, GitHub metadata, Product Hunt activity, Hacker News discussion, YouTube coverage, manual imports, and other compliant sources as they are added. Restricted sources stay parked until there is a reliable, compliant access path.
Why Scores Change
A score can move when a tool ships a major update, changes pricing, earns stronger third-party evidence, gets fresh community attention, or becomes a better/worse fit for a refined use case. The What Moved page and live ticker are the public trail for those changes.
How To Read Low Evidence
Low Evidence Level does not always mean a tool is bad. It means the recommendation has fewer receipts behind it today. In that case, keep comparing, open the evidence drawer, and treat the score as directional rather than definitive.