Retail

See what your locations are actually doing.

The dashboard says the promotion is live in 200 stores. The dashboard is describing a plan, not a shelf.

The problem

Compliance is reported by the same people being measured, and the gap between what was scheduled and what is physically on the floor stays invisible until a customer finds it.

What happens today

Field managers drive a route and photograph what they remember to. Coverage is thin, the data arrives late, and nothing in it can be checked.

The standard jobs

Named, versioned, and verifiable because they were stated in advance.

RETAIL_LOCATION_CHECKIs the location where it should be, and does it look like a going concern?
BUSINESS_OPEN_CHECKOpen at the hours claimed?
RETAIL_SIGNAGE_CHECKIs the signage up, correct and undamaged?
RETAIL_SHELF_CHECKIs the product on the shelf, faced, and priced as intended?

What comes back

Evidence, not a photograph somebody has to interpret.

Place and time are deterministic and gate the payout. The model's reading of the photographs is advisory: it can permit auto-approval or route to a human, and that is all. No verdict at all is never a pass.

One call

Request
POST /v1/tags/recipe

{
  "recipe": "RETAIL_SHELF_CHECK",
  "location": { "lat": 33.749, "lng": -84.388 }
}

Run one location to try it, or a campaign across every store in a region. A campaign of 200 locations is 200 ordinary Tags — combining them would be easier, and would destroy the only thing worth paying for.

What it costs

You set the bounty. We add 20%.

A $75 Tag costs you $90 and pays the Partner $75. Nothing is charged for a Tag nobody claimed, for proof that failed the place or time lock, or for work you disputed until a human has ruled on it.

Full pricing · API documentation

Throw one and see.