Words, visuals and stories, not numbers
A star flattens the story. Gist turns what real people felt — in their own words — into one honest index you can read in seconds. For everything worth an opinion, not just films and restaurants.
Films score out of 10, books out of 5, apps out of 100 — none comparable, all gameable. And “4 out of 5” means something completely different at a corner café than at a tasting menu. Not one of them answers the only question you came with: should I?
The math was broken from the start
A star is an ordered word — “okay,” “good,” “great” — wearing a number's costume. The gaps between them aren't equal, so averaging them is statistically meaningless (Stevens, Levels of Measurement, 1946). Two things can post the same 3.0 and mean the opposite:
Gist never averages. Every worded signal is weighed by how much it should count — and when people disagree, the split is shown, not smoothed away.
Gist is built on four ideas that don't exist anywhere else — together they turn a lifetime of vague ratings into something you can actually act on.
Instead of guessing a number, you pick the words that fit — “Outstanding, best I've had,” “A bit steep,” “Stayed with me for days.” Every phrase carries a precise, weighted meaning. Human on the surface, exact underneath.
Everywhere else, one account equals one vote — so a bot farm equals a thousand people. On Gist, weight is earned: reviewers build per-category reputation over time, and fuller, more helpful reviews move the index more. A drive-by star from a day-old account barely registers — and a behavior-based integrity layer watches for the brigades that try anyway.
Every worded, weighted, verified signal rolls into the Gist Rating Index — a single multi-dimensional score, not a naïve average. You get the whole truth in one glance, and the detail underneath whenever you want it.
Most platforms quietly reweight the score for who showed up, and hope you never ask how. Gist does the opposite: it shows you the mix of people behind the number, and lets you filter to people like you. We never bend the score to “correct” a crowd — we make it legible and hand you the controls. Opt-in, and yours to ignore.
No 1-to-10 sliders. No essay you'll never finish. Say how it landed, tap the words that fit, drop a photo if you've got one. Thirty seconds or five minutes — the depth is yours, and every choice becomes signal.
A peek at the flow — the Gist rating scale itself is still in design.
A five-star rating can't tell you why someone picked a thing, what they expected, or whether it delivered. Gist asks — quietly, optionally — so a review carries the context that actually predicts your experience.
Words, photos, the emoji that just fits, a mood you can tap — capture the experience the way it actually happened. Gist reads your words for sentiment and folds every signal into the same honest index.
Thousands of words are more than anyone can read. Gist reads them for you — ask a real question, get an answer drawn only from the reviews, with the receipts. No sponsored nudges, no guessing.
Built on a multi-model engine with graceful fallback. Answers switch on at launch.
Reviews shouldn't be a one-way court. Claim your entity, reply on the record, and — when an old brigade or one bad week has fossilized your score — request rehabilitation or appeal a coordinated attack. Built in, with the audit trail to back it.
One review language, one honest index — across every kind of thing worth an opinion. Eight categories at launch, on a single scale built so the next one is just data, not a rebuild.
One honest scale for everything worth an opinion — coming soon.