How Our Community Reviews Work
You don't write a review here — you build one, tap by tap, and the sum becomes the rating no one can buy.
The Leaf Guide
Original cover art — The Leaf Guide
Ask most people to review an eighth of flower and you get one of two things: a blank text box they'll never fill, or five stars and the word "fire." Neither tells the next shopper anything. The blank box loses the nuance because writing is work; the star-and-emoji loses it because the format has no room for nuance to live. Both failures have the same downstream effect — a rating you can't trust, because you can't see what's underneath it.
We built the guided review to fix the format, not to nag you into writing more.
You tap; it drafts
Open a product page and you won't find an empty field waiting for prose. You'll find a short sequence of choices. Effect — where did this actually take you, head or body, up or down, and how hard. Flavor — the notes you caught, from gas and citrus to earth and pepper. Method — how you took it in, because a joint, a dry-herb vape at 385°F, and a dab are three different experiences of the same jar. Value — what you paid against what you got. Rebuy — the single most honest question in retail: would you spend your own money on this again. And the leaf rating, our overall mark.
You tap through those. The tool assembles them into a clean, readable draft in plain language — a paragraph that reads like something a thoughtful person wrote, because the structure did the writing for you. Then, if you want, you add a personal comment: the detail no menu can anticipate. The couch you didn't leave. The batch that hit softer than the last. The thing that made you laugh. That comment is yours; everything above it is data.
This matters because the tap-through isn't a shortcut to a paragraph — the paragraph is the byproduct. The structured choices underneath are what the community actually reads.
Structured data is the whole game
When you select "citrus" instead of typing the word, "citrus" becomes something the system can count. Multiply that across a few dozen contributors and a jar's flavor stops being one person's adjective and becomes a distribution — a shape the community drew together. Free-text can't do that. "Smooth," "chill," "clean," and "mellow" might all mean the same experience, but to software they're four unrelated strings. Structure is what turns scattered impressions into a signal you can actually read at a glance.
It also lets us hold onto something the industry throws away constantly: terroir. In our sense that's the full fingerprint — strain by grower by brand by year by batch. The same cultivar from two farms is two products. The same farm's Gelato in a wet fall and a dry one is two products. And the batch that scored beautifully in March can drift by August, because living agriculture drifts. Capture the review as structured fields tied to that specific fingerprint and the drift becomes visible instead of invisible. Tying feedback to a specific lot is an approach the more rigorous corners of the category have started to adopt — see Databis tying reviews to specific batches, and Leafly's tasting standards formalizing the sensory vocabulary. A plain star widget flattens all of that into one number and calls it a day.
This is the rating system
Worth saying plainly, because it's easy to skim past: the guided review isn't a feature bolted onto a star rating. It is the rating. The leaf mark you see up top is the community's structured input, aggregated — not an editor's decree. What the room said is what shows.
That's a deliberate design choice with a cost. It means a product with three thoughtful reviews carries less certainty than one with sixty, and we'd rather show you that thinness than paper over it with a confident-looking average. Fewer data points, less confidence. That's not a bug we're apologizing for; it's the honest reading of a young sample, and you deserve to see it.
Community-decided, never pay-to-play
Here's the line we won't cross. No brand can buy a rating, reorder a page, bury a bad batch, or purchase placement in a result. There is no ad tier that quietly outranks an organic one, and no separate score living behind a curtain. The failure mode we're designing against is the oldest one in reviews — the incentive to sell the thing you're supposed to be judging — and the defense is structural: the score is generated from community taps we don't sell access to, full stop. We'd rather remove the temptation from our own architecture than trust ourselves to resist it.
So what does move influence around? Quality of contribution. You can upvote reviews that helped you — the batch note that flagged a change, the method detail that explained why your experience differed, the comment that saved you thirty dollars. Upvotes surface the useful and let the low-effort settle. The currency here is being genuinely helpful to the next person in the aisle, and it's the only currency we mint.
Why we bother
There's a plainer version of all this. A rating is only worth the trust behind it, and trust is earned in the boring accumulation of honest, structured, batch-aware contributions from people with nothing to sell. The guided review is simply the fastest way we've found to let you add to that pile in ninety seconds — so that when you look something up before you spend, you're reading the room, not the marketing.
Tap through one this week.
Nothing here is medical advice. Content is for adults 21+ and nothing is for sale.
Sources: Databis — Verified Cannabis Product Reviews by Batch, Leafly's Cannabis Rating System & Tasting Standards