AisleWatch AI

AisleWatch AI / Loss prevention / Shoplifting detection camera

Shoplifting detection AI that catches theft before the shopper leaves.

Most shop theft is found weeks later at the stock count. AisleWatch AI watches your high-loss shelves and sends staff a short clip while the shopper is still inside. Less stock walks out of the door.

Shoplifting detection camera view of a store aisle with the high-loss bay marked for review and a shopper tracked, face blurred
Simulation.

AI theft detection retail teams can act on

Store teams cannot watch every aisle, and footage pulled after the count rarely gets anything back. AI theft detection in retail helps only when staff hear in time to walk over.

93% rise in the average number of shoplifting incidents reported by US retailers, 2023 against 2019. Source: NRF, The Impact of Retail Theft and Violence 2024 (third-party public figure)

What staff see when an item goes into a bag

The nearest staff member gets the aisle and a short clip as the alert is raised. Nobody has to search the footage later.

  • An item put into a bag or clothing.
  • Several items swept off a shelf.
  • Someone in the store after closing.
  • A known repeat offender, on the optional watchlist.

A store team member checks every clip before anyone speaks to a shopper.

Shoplifting detection camera still: a shopper with a basket beside the high-loss bay marked for review, face blurredSimulation.

Real-time alert16:41:12

A shopper may have hidden an item at the high-loss bay.

Aisle 4, Store 014

Sent to
Store team, Store 014
Channels
Microsoft Teams, SMS, email and the dashboard
  1. 16:41Store team on the floor, by Teams
  2. 16:43Store manager, if nobody responds
  3. 16:46Area manager, if still open

Shoplifting detection on existing CCTV, step by step

  1. The behaviour is flagged

    AisleWatch follows hands, items and bags in the marked bay.

  2. A person reviews

    A staff member watches the clip and decides whether to walk over and offer help.

  3. The outcome is logged

    No action, a staff member in the aisle or a sale saved is kept with the clip.

  4. False alerts are counted

    Each alert marked false is sent to loss prevention every week of the pilot.

How the shelves are set up

  1. High-loss bays are marked

    Fragrance, spirits, razors or electronics are marked on the camera views you have.

  2. Alerts are tuned per store

    Each store's cameras and bays get their own settings, and the alert rate is reported weekly.

  3. Cases are kept

    If you report a case, the clips and times are ready for the police or your insurer.

Case 0412Store 014 ยท aisle 4, high-loss bay
  1. Item possibly hidden, sent for review
  2. Store team on the floor alerted, by Teams
  3. Store manager acknowledged
  4. Clip reviewed at aisle 4
  5. Case logged with 2 clips
Clips
0:38 from two camera angles
Outcome
Item paid for at the till
Kept
30 days, then deleted
Simulation.

Questions about shoplifting detection

Send the AisleWatch team another question

How does shoplifting detection AI work?

Shoplifting detection AI follows hands, items and bags on the cameras over your shelves. When an item looks hidden, a staff member gets a short clip to review.

Can AI detect shoplifting on the CCTV we have?

AisleWatch runs on most IP cameras with a clear view of the shelves. False alerts are reported to loss prevention each week, so you can see how it performs in your stores.

Does AisleWatch use face recognition?

Faces are blurred in shared clips by default. A watchlist of known offenders is an option that only the retailer can switch on.

What happens after a shoplifting alert?

A staff member watches the clip and chooses the response, such as standing near the aisle. AisleWatch records what happened and never accuses anyone.

Is shoplifting detection offered in Singapore and the Gulf?

AisleWatch offers it to retailers in Singapore, Saudi Arabia and the UAE. Gulf work is delivered with local partners, and pilots run for 8 weeks in one or two stores.

See a sample store report from one day of your own footage.

AisleWatch returns the alerts one day of your footage would have raised. The footage is covered by an NDA and deleted after review.