Strain matching built from your shop’s reviews.
Every cannabis POS has a strain catalog. Most of them feed you the same Leafly metadata everyone else has. CannAgent builds your strain intelligence from your own customer reviews— what your shop’s buyers actually said about each strain after they tried it.
Your reviews train the recommendation engine.
Customer buys a strain, leaves a review (4 hours later via the review-collection surface). The system reads the review text against the strain catalog — pulls out effect signals, terpene-association signals, comparative-strain signals.
Over time, your shop’s strain catalog accumulates a profile per strain that reflects what your customers actually experience, not what a national catalog scraped from somewhere. Then when a new customer walks in and says “I liked GG4 but I want something less heavy,” the budtender gets a ranked list of strains your shop carries that your reviewers describe as similar.
Scoring internals: <Redacted> — the model is part of the patent posture; the output is what ships to the budtender at the cart.
What the engine extracts is review pattern data — words your reviewers used about a strain. What the budtender SEES at the cart is a ranked similar-strain list with your-customers-described-it tags. What the customer never sees is an effect claim made by your shop — WAC 314-55-155 stays clean because the language at the cart is comparative (“similar to what other customers who liked X also came back for”), never therapeutic.
Two flowers labeled the same brand don’t hit the same.
Your customers know this. Two pre-rolls labeled “Blue Dream” from two different cultivators don’t smoke the same. A national catalog can’t tell you which version your shop’s repeat buyers actually came back for.
Strain intelligence from your own reviews surfaces the recommendation that matches the experience your shop delivers. Budtenders stop guessing. Repeat-visit customers feel known.
No efficacy claims. No medical recommendations. This is pattern-matching on what your customers reported, not a doctor’s note.
See it on your data.
Bundled with Co-Pilot at $699/store/mo solo, $599/store/mo multi-store. The 30-minute demo loads your shop’s review corpus and walks the strain-matching output.