Visual merchandising
Every display, the way the VM guide shows it.
Bays, windows and focus displays photographed at every store, and checked before the photo is sent.
The problem
The VM guide went to 200 stores. How many look like it?
Store photos land in a WhatsApp group and nobody checks them all. Ops AI checks each one as it's taken.
Is the window complete?
The checklist says:
A photo in the groupOps AI checks: Window checked: complete, lit, no torn or crooked graphics
Is the bay full?
The checklist says:
Ticked: yesOps AI checks: Gaps spotted in the bay photo, facings counted
Is the right planogram up?
The checklist says:
Ticked: yesOps AI checks: The planogram's code read from its sheet or screen
Is the focus display built?
The checklist says:
Ticked: yesOps AI checks: Display photo checked: fully stocked, campaign signage on
Is the signage current?
The checklist says:
Ticked: yesOps AI checks: Torn, faded or ended-offer signage photographed when it's wrong
The size of the problem · Published research, not our numbers. Each one has its source.
8.3%
of grocery shelf items were out of stock on an average day, in a 2002 worldwide study
4%
of sales a typical retailer lost to empty shelves, in the same study
Ready to use
VM checklists that ship with Ops AI.
Start from the library and say exactly what each photo must show.
Planogram and VM check
One bay against its planogram: placement, facings, tags, presentation.
Reads the code · counts facings · spots gaps
Campaign launch and window
Window, entrance and focus display set up on launch day.
Checks the window and the display
Shelf availability walk
One aisle walked for empty shelf spaces.
Counts empty spaces
Monthly store audit
Sales floor, presentation and planogram, scored section by section.
Spots gaps on the floor photos
For VM and area managers
See which stores need re-merchandising, before you visit.
Every store side by side
Launch and VM checks done on time, by store, zone and day.
The photo behind every answer
Open any store's check and see the bay or window as it was.
One launch, every store
Send the campaign checklist to every store with a due time.
Head office
Ask which stores aren't launch-ready.
Ask in plain English or Hindi; the answer comes with the numbers behind it.
Answered from your reports
Which stores didn't finish opening checks before 10?
2 of 6 stores finished opening checks late this week. Store 105 was late on 3 days.
- Store 105
- Late on 3 of 7 days
- Store 103
- Late on 1 of 7 days
2 numbers checked against checklist resultsOpen checklist results
Draft broadcast
Send a reminder to the 2 store managers, with acknowledgement required.
Questions
Quick answers.
What the AI doesn't do
It doesn't compare the photo with a planogram image or your campaign brief, and it doesn't judge product placement: staff answer that. It checks each photo against what you say it must show, counts facings and spots gaps.
Can it check our displays against our VM guidelines?
Yes, in words. Each photo question says what the display must show, and the AI checks every photo against it as it's taken. It doesn't compare against a reference image.
What happens when a display fails?
You choose: block the photo until a retake passes, warn and ask for a reason, or accept it and flag it for review.
Can we send a new campaign to every store at once?
Yes. Send the launch checklist to every store, a region or chosen stores, with a due time, and see which are done.
Does it stop staff reusing old photos?
Photos come from the in-app camera unless a question allows the gallery, and every one is stamped with the time and place.
Bring one VM guide. See which stores match it.
We'll set up your display checks and show you what the photos catch, store by store.
- Access enforced in the database
- Every change logged, before and after
- Every number traced to its photo
- Security

