Brands
brand × marketDistribution, doors carried, coverage, category mix, and the SKUs a brand has on shelf.
Shelf Data API
Every dispensary shelf, as structured data.
Query who carries what, where, at what price, in stock or not — and what changed — across the cannabis retail market. A clean, read-only REST API over the same shelf intelligence that powers the CannaiQ dashboards. Build it into your own app, model, or warehouse.
10,000+dispensaries observed
56markets · US · CA · PR
18,000+brands tracked
7M+SKUs observed
< 1 hrrefresh on live stores
Six things you can pull
Six resources, each at a defined grain. Filter, combine, and expand them into the exact slice you need.
Stores
dispensaryStore directory, location, platform, and the brands and full menu each store carries right now.
Products
product familyCanonical product families deduped across stores — one identity per real product.
SKUs
sellable variantVariant-level catalog — weight, dose, strain, format — and every store carrying each SKU.
Inventory
SKU × store, latest readIn-stock / out-of-stock and quantity from the most recent shelf read. Plus what just left the shelf.
Signals
change event, windowedNew placements, lost placements, price moves, category trends, and movement direction over time.
Every number carries its receipts
No bare numbers. Each value ships with where it came from and when we last saw it — so you always know whether you're reading a measured fact or a modeled estimate.
observed
Read directly from a shelf: presence, price, promotions, stock, door counts. A measured fact.
inferred
Computed from change between shelf reads: movement, velocity, trends. Always directional, always a lower bound — labeled as such.
confirmed
Observed data upgraded by a brand's own catalog verification. The highest-confidence tier.
The endpoints
A deliberately lean REST surface — depth comes from query flexibility, not endpoint sprawl.
Brands
GET /data/v1/brandsGET /data/v1/brands/{id}GET /data/v1/brands/{id}/storesGET /data/v1/brands/{id}/skusGET /data/v1/brands/{id}/coverageGET /data/v1/brands/{id}/categories
Stores
GET /data/v1/storesGET /data/v1/stores/{id}GET /data/v1/stores/{id}/brandsGET /data/v1/stores/{id}/menuGET /data/v1/stores/{id}/categories
Products & SKUs
GET /data/v1/productsGET /data/v1/products/{id}GET /data/v1/skusGET /data/v1/skus/{id}GET /data/v1/skus/{id}/shelf
Inventory
GET /data/v1/inventory/stockGET /data/v1/inventory/oosGET /data/v1/inventory/delisted
Pricing
GET /data/v1/pricing/skusGET /data/v1/pricing/ladderGET /data/v1/pricing/market
Signals
GET /data/v1/signals/new-placementsGET /data/v1/signals/lost-placementsGET /data/v1/signals/price-changesGET /data/v1/signals/velocityGET /data/v1/signals/category-trends
One predictable shape
Every response returns data, a meta block with provenance,
timestamp, grain, and coverage, and a page block for pagination.
Inventory is always the latest shelf read; trends are windowed and labeled.
- SKU counts are distinct SKUs — never inflated by store-by-store listings.
- Inventory is point-in-time — the most recent read, not a rolling average.
- Trends grow with history — windowed signals are honest lower bounds.
{
"data": [
{
"sku_id": "…",
"name": "Green Apple Pocket Spritz 100mg",
"brand": "Boho Euphorics",
"format": "beverage",
"store_id": 2498,
"store_name": "Authorized Dealer (Willingboro)",
"city": "Willingboro",
"state": "NJ",
"in_stock": false,
"list_price": 24.00,
"last_seen_at": "2026-06-23T14:02:00Z"
}
],
"meta": {
"as_of": "2026-06-23T14:02:00Z",
"provenance": "observed",
"grain": "placement",
"window": "snapshot",
"coverage": { "states": ["NJ"], "confidence": "high" }
},
"page": { "limit": 100, "offset": 0, "total": 1 }
}Slice it your way
Every list endpoint speaks one query grammar — so you compose the view you need without us pre-building it.
Filter
Combine market, brand, category, platform, price range, stock status, and "changed since" — e.g. ?state=NJ&category=flower&price[gte]=20&price[lte]=40&in_stock=true.
Select fields
Return only the fields you want with ?fields= — smaller, faster, cheaper payloads.
Sort & page
Sort on any indexed field (?sort=-doors), page with limit/offset or cursors for large pulls.
Expand
Walk relationships inline — a SKU to every store carrying it, a brand to its full catalog — in one request with ?expand=.
Bulk export
Enterprise plans can export any resource as Parquet or JSONL, scoped to your markets, to load into your own warehouse.
Scoped & metered
Keys are scoped to your markets and resources, rate-limited, and metered per call — predictable and auditable.
We never sell a number we can't see.
Everything in the API is derived from observed dispensary shelves — not modeled point-of-sale figures, not foot-traffic estimates, not survey panels. We don't fabricate retail sales, basket data, or anything we can't actually observe. Where a value is an estimate, it says so, in the data. That honesty is the product.