Faceless depth sensors over billing counters that say when to open another till
TL;DR Tillpace mounts LiDAR or depth sensors above the billing zone of hypermarkets and large supermarkets. It counts people in each line without capturing faces and pings the floor manager when waits are about to cross a set limit. Chains pay a monthly fee per store for shorter queues on the busiest days.
The worst hour in an Indian hypermarket is a Sunday evening before Diwali, when billing lines snake right into the aisles. Managers open counters by eyeballing the floor, usually after the line is already long. Shoppers with a half-full trolley leave it and walk out, and that lost sale never shows up in any report.
Store numbers keep climbing. The largest retail conglomerate alone ran about 19,340 stores in February 2026, and retail sales grew 10% in August 2026 according to an RAI survey[1].
Cameras could watch the queue, and one Indian vendor already sells checkout queue alerts in India[2]. But plenty of chains and shoppers are uneasy about face-capable video being used for analytics, and the DPDP Act treats that footage as personal data[3].
The seed pictured LiDAR heatmaps across the whole store, and we changed that. Covering an entire hypermarket with LiDAR is expensive (vendors don't publish prices), and a heatmap of the whole floor rarely changes a decision. The billing zone is where a live number pays for itself, so Tillpace covers only that.
Two to six ceiling sensors watch the counters. Software counts people per line, tracks how fast each line moves and forecasts the wait for the next shopper. When the forecast crosses the store's threshold, say seven minutes, the floor manager's phone buzzes with "open counter 9". After the rush the system reports how many minutes each counter sat over the limit and how many staff hours went in.
LiDAR sees shapes and distances, not faces. One lidar maker markets its software as anonymised by design, with no faces or licence plates[4]. That's the vendor's claim though, so Tillpace should still test it and keep no raw point clouds longer than it needs to.
The buyer is the operations head of a hypermarket or large-supermarket chain, where billing staff are already the biggest daily cost decision. My working assumption is ₹6,000–12,000 per store per month on a three-year contract, hardware included. Mall operators are a second buyer for food-court and parking-pay queues, where the same sensors do the same job.
Depth and LiDAR sensors now ship with ready people-tracking software, so the hard perception work is off the shelf[4]. Retail space leasing hit a record 8.9 million sq ft in the top seven cities in 2025[1], which means more large-format stores. And the DPDP Act makes camera analytics costlier, which suits a sensor that never sees a face[3].
There is no public figure for India's checkout-queue analytics market, and no vendor publishes Indian sensor prices. The narrow signal is the number of large-format stores at the biggest chains, the largest of which runs about 19,340 stores of all formats[1].
Only a fraction of those are hypermarkets. Treat the total as an upper bound and not as your market.
One Swiss firm sells 3D people-counting sensors with real-time queue detection to airports and retailers[5]. An Indian vendor does camera-based checkout queue alerts alongside footfall and loss prevention[2]. Lidar makers supply the lidar and tracking software but not a retail product[4].
The gap is an Indian-priced, faceless system sold only for the billing zone, with a staffing report managers can act on.
| Difficulty | Moderate (47/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 9–12 months |
The hardest parts are sensor cost per store, reliable counts in dense, chaotic Indian billing areas full of trolleys and kids, and installs across many cities. Proving lost sales from abandoned trolleys is tough too, and that drags out the return-on-investment pitch.
There's no licence to run queue analytics, but shipping hardware brings a few approvals.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI counts blindly in a new store. Check them by hand during a pilot week, and don't rely on the vendor's claim alone that point clouds stay anonymous.
Want to work out the AI layer for Tillpace? Talk to Yash →
Here's what can go south -
I think this is worth a shot for a founder with sensor or IoT hardware experience and contacts in large-format retail. A software-only team should skip it. Prove first that a pilot store cuts minutes over its wait limit by a third during a festival week, using only existing staff.
Plan for roughly ₹1.2–2.2 Cr, with sensors, installs across cities and software development taking most of it.
No licence to run the analytics, but a hub with Wi-Fi or Bluetooth needs WPC type approval, and cameras or adaptors may need BIS registration.
Yes, large stores everywhere deal with billing queues. Check local radio equipment approvals and privacy law before you ship sensors.
The free Tillpace validation kit has the line-by-line capital breakdown, the effort scoring, the radio and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for store operators, and 5 assumptions to test. Download the kit →
Thinking about building Tillpace or something close to it? The make-or-break work is getting sensor cost per store low enough that the monthly fee still makes sense, 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.