Fixed shelf cameras that tell regional supermarket staff which bay to refill
TL;DR Gapspot puts a few cheap fixed cameras over a supermarket's best-selling bays and sends floor staff a refill task when a facing goes empty. It sells to regional chains with 10 to 60 stores, which pay a monthly fee per bay. The pitch is recovered sales, not another dashboard.
An empty shelf is a lost sale, plain and simple. Across retail studies the average out-of-stock rate is 8.3%, and when shoppers hit a gap, 31% buy the item at another store and 9% buy nothing[1]. Often the stock is sitting in the back room and just didn't reach the shelf in time.
Indian regional supermarket chains feel this hard on weekends and festival days, when a handful of staple and snack bays empty fast. Brand merchandisers audit shelves, but they come on the brand's schedule and report to the brand.
So the store manager usually finds the gap when a customer complains. We found no Indian figure for shelf gaps in modern trade.
Gapspot covers the 30 to 80 bays that drive most of a store's sales, not the whole store. Small cameras mounted on shelf edges or the ceiling take a frame every few minutes. A vision model compares each bay with its expected layout and flags empty or thin facings.
When a gap shows up, the floor staffer for that aisle gets a task on a shared phone with a photo, the bay and the product. The task closes when the next frame shows the shelf full again. Managers see how long gaps lasted, by shift and by aisle, and which products had run out in the back room too.
Cameras point at shelves. People caught in a frame get blurred on the device and only shelf crops leave the store, which keeps the privacy question small and the bandwidth cheap.
The product is narrow on purpose. Planogram audits and share-of-shelf reports already exist for brands. Gapspot is a refill tool for the retailer's own staff.
The regional chain's operations head pays a monthly fee per bay covered, with cameras leased into the fee, and store staff use it. The price has to sit well below the margin on the sales it recovers. We found no public price list for shelf-monitoring services in India, so set it from pilot data. A later line is selling anonymised gap data to brands, with the retailer's consent and a revenue share.
Photo-based shelf recognition now works at scale in India. One vendor reports 98% SKU-level availability accuracy in one deployment[2], and another claims over 97% accuracy across 400,000 stores a month[3]. So the models are proven. What hasn't changed is who they serve, and that's mostly brands.
There is no public figure for what Indian retailers spend on shelf monitoring. You can see the buyer pool though. National chains are growing fast, with the biggest value supermarket chain passing 500 stores in April 2026[4], and regional chains compete with them on availability.
Adjacent - IBEF puts Indian retail at US$1,093.89 billion in 2025 (about ₹105.8 lakh Cr)[5], which is context and not your market.
Shelf-recognition vendors with Indian teams analyse shelf photos taken by merchandisers and sell mostly to FMCG brands[2][3]. Their model depends on a person visiting the store with a phone, so the data turns up after the gap has already cost sales. Large chains can build their own tools.
The gap is a fixed-camera refill system priced for a 20-store regional chain with no data science team.
| Difficulty | Moderate (50/100) |
| Starting capital | ₹1.04–2.05 Cr |
| Time to launch | 9–12 months |
The model is the easier half. Mounting cameras that survive cleaning, glare and restocking in real stores is harder. Getting staff to act on alerts during rush hours is the hardest of all, because that's a change in store routine and no software fixes it on its own.
No sector licence applies to shelf monitoring, but the cameras and footage need care.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust the model blindly on new packaging or odd lighting. Check a sample of alerts by hand every week, because false gaps waste staff time fast.
Want to work out the AI layer for Gapspot? Talk to Yash →
Here's what can go south -
I'd back a founder who has run store operations and can sell to regional chain owners face to face. If your team wants to sell pure software from a distance then skip this one. Prove one thing first in a pilot store - covered bays show at least 30% shorter gaps than uncovered bays over four weeks.
Plan for roughly ₹1.04–2.05 Cr, mostly on cameras, installation, model training and field teams. Covering only the top bays in one pilot store keeps early hardware costs down.
No sector licence. Buy BIS-registered cameras, and follow the DPDP Act by blurring people and keeping only shelf images.
Yes, supermarkets everywhere lose sales to empty shelves. Check local camera certification and privacy rules on in-store video first.
The free Gapspot validation kit has the line-by-line capital breakdown, the effort scoring, the camera and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for store managers and chain owners, and 5 assumptions to test. Download the kit →
Thinking about building Gapspot or something close to it? The make-or-break work is proving a sales lift that a regional chain owner believes, and that's exactly the kind of thing I help founders with. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.