Where is everything?
Manual shelf audits are slow. The layout changes again before the report travels back to head office.
Solution / Retail Planogram
Get your shelves right.
On-prem computer vision and analytics that turn a shelf photo into a structured layout. See which row, facing and product is working, then decide what to change.
Discuss your shelvesSee how it works
The floor and the numbers don't agree.
The shelf is a blind spot.
POS tells you a product sold. It doesn't tell you which row held it or whether the planned facing was there.
Data without a decision.
A store-wide number does not explain which shelf needs attention this morning.
From a photo to a decision.
Capture
Photograph a shelf with a phone. No special capture rig.
Detect + review
Match products to the catalog. A person confirms the uncertain ones.
Connect
Tie the structured shelf to sales, down to the row and product.
Act
Work from a prioritized list of exceptions, not an undirected spreadsheet.
See the store as work you can act on.
The store, prioritized.
A morning list surfaces empty slots, top sellers that stalled and shelves overdue for a capture. Rank attention by revenue at risk.
What needs attention
Empty facingCheck the floor
Stalled productReview stock + placement
Old capturePhotograph shelf again
Every shelf, as a report.
Revenue, products, facings, a 30-day trend and revenue by weekday, in one shelf report. Export a PDF for store operations.
What a shelf report contains
ShelfRow → Product → Facing
Sales + captured layout
Trend + weekday view
Review → PDF → Store ops
Sales + captured layout
Trend + weekday view
Review → PDF → Store ops
The sales, explained.
Category share, daily revenue and products gaining or losing momentum. Read the individual SKU rather than only the whole department.
From the store to the shelf
StoreShelfRowProductFacing
Increasing resolution, not a claim of increasing revenue.
Shelves
Layout, capture history, facings and catalog gaps in one place.
Products
What moves, what stalls, and where each SKU sits on the floor.
Stores
Capture coverage and revenue across locations.
04 / On-prem
Your data stays in the building.
The deployment runs locally on retailer hardware. Shelf photos and POS data stay on-site, with no vendor-controlled cloud required.
Local data
Keep capture and sales together inside your own infrastructure.
Your model
You own the model, rather than paying per seat, store or call.
Infrastructure, not a slogan
Hardware, installation and operation still have costs. Scope those against the actual store environment.
05 / How it compares
From recording shelves to fixing them.
The jobManual auditRetail Planogram
Record the shelfCount products and write down their positions.Take a photo, check the matches and confirm the layout.
Find the gapsCompare notes with the planned shelf.Review differences between the planned and actual shelf.
Understand salesLook up sales in a separate report.Read sales beside the shelf, row and product.
Choose the next jobWork through reports to spot exceptions.Start with the empty space, stalled products and shelves needing a new photo.
Built with a supermarket chain.
A large retailer in Southeast Asia runs this today. The client is unnamed. Here is exactly what that deployment shows, and what we do not claim.
Case fileIn operation
- Client
- Supermarket chain, Southeast Asia (name withheld)
- Runs on
- On retailer hardware, on-prem
- Capture
- Phone photos, confirmed by a person
- Visibility
- Shelf, row and slot level
- Output
- Prioritized exceptions and per-shelf PDF reports
What is evidenced
Shelf, row and slot-level visibility from a phone photo, with a human check on every uncertain match.
What we do not claim
No verified before/after dataset is published. Public demo dashboards use fictional stores and illustrative figures, not customer results.
Case study▶
Retail Planogram: from shelf photo to decision
How a supermarket chain moved from a shelf photo to a prioritized job.
Read the case studyTell us about your stores.
Start with the shelves, catalog and operational question you need to answer. Then discuss a walkthrough, a bounded pilot and the on-prem deployment requirements.
Discuss Retail PlanogramOr write to founders@generalmachines.ai