GST and money
UPI Passed Cash at an Optical Counter in 2026: 2,883 Bills, Four Years
In 2022, six bills in ten were cash. In 2026 cash is under four in ten and UPI is the majority method for the first time.
In 2022, six bills in ten at this counter were settled in cash. In 2026, cash is under four in ten and UPI is the majority payment method for the first time.
That is not a survey or an industry forecast. It is 2,883 bills raised at one Indian optical shop between 2022 and September 2026, counted by how each one was paid.
Disclosure and limits. I build Optocore, an optical ERP, and these numbers come out of it. They are one Indian optical business with two branches — 2,883 orders with a payment method recorded, 2022 to 22 September 2026. This is a real shop, not a benchmark. No amounts are published here, only how many bills were paid which way.
What changed, year by year
Every bill in this system records exactly one payment method. The shares:
| Year | Bills | Cash | UPI | Card |
|---|---|---|---|---|
| 2022 | 111 | 58.6% | 29.7% | 11.7% |
| 2023 | 891 | 49.5% | 38.9% | 11.6% |
| 2024 | 822 | 47.1% | 43.4% | 9.5% |
| 2025 | 753 | 47.3% | 45.8% | 6.9% |
| 2026 (to Sep) | 306 | 37.3% | 52.0% | 10.8% |
Three things are worth reading out of that table, and only one of them is the headline.
Cash lost 21 points in four years. Not to a collapse — to a steady, boring slide of about five points a year. Nothing in the shop changed to cause it. The customers changed.
UPI did not take those customers from card. Card sat between 9% and 12% the whole time, dipped to 6.9% in 2025, and came back. The card machine is not being replaced; the cash drawer is.
2025 looks like a pause and was not. Cash actually ticked up a fraction that year while UPI kept climbing, because card gave up ground to both. A shop watching only cash-versus-UPI would have concluded the shift had stalled, twelve months before it crossed over.
Why this matters at the counter, not in a report
An optical shop is not a kirana counter. The average spectacle sale is large enough that the payment method decides how the rest of the day works.
The drawer stopped being the till. When six bills in ten were cash, counting the drawer at close was very nearly counting the day. At four in ten it is not even half the answer, and a day close that reconciles only cash is reconciling the smaller part of the shop. What has to be checked now is cash and the UPI settlement and the card batch, against what the system says was billed.
UPI arrives on a different clock. Cash is in the drawer the moment the customer leaves. A UPI collection lands in the bank on the payment service provider's schedule, and a card batch settles later still, net of charges. Three tenders, three timings, one day's sales — which is exactly why "the bank balance does not match the sales report" is a question that started being asked in optical shops around 2024.
Refunds and returns follow the tender. A cash refund is a decision. A UPI refund is a reversal that has to be traceable to the original collection, and a card refund goes back to the card. If the bill does not record which method paid it, a return three weeks later becomes an argument.
What a shop should actually do about it
Nothing here requires new software to understand, so it is worth separating the observation from the sales pitch.
- Record the method on the bill, every bill. Not in a notebook, on the invoice. It is the only thing that makes a day close, a refund and a dispute resolvable later. Our own bill format reference sets out where it goes.
- Close the day by method, not in total. Cash counted, UPI checked against the app, card against the batch. A single total hides exactly the error you are closing the day to find.
- Expect the crossover to keep going. At roughly five points a year, cash is heading for a third of bills and then a quarter. A counter still built around a cash drawer is being designed for the shop as it was in 2022.
Three questions for whoever sold you your billing software
None of this needs a new system to notice. It does need a system that can answer:
- Does it record the method on the bill, and on each part-payment separately? An advance in cash and a balance by UPI are two facts, not one.
- Does the day close split by method? Cash counted, UPI against the app, card against the batch — each against what was billed, not one total.
- Does a refund know how the money originally came in? Three weeks later, that is the difference between a reversal and an argument.
If the answer to any of them is "we manage it in a register", that register is now tracking the majority of your takings.
What these numbers cannot tell you
This is the honest part, and the part that will be missing from every other article on this subject.
No bill in this data records a split payment. Not one of the 2,883. The system these bills came from accepted a single method per bill, so a customer who paid part in cash and the rest by UPI was written down as one or the other, by whoever was standing at the counter. Given that 46.8% of these bills were part-paid — an advance at order, the balance on collection — the real number of split-tender sales is certainly large and is simply not recorded anywhere.
So the shares above are the shares as recorded, which for a single-tender bill is exact and for a part-paid one is the tender somebody chose to write down. That is a limit of the register, not of the shop, and it is worth knowing before anybody plans around a percentage.
Nothing here says anything about amounts. A payment method count treats a ₹400 lens solution and a pair of progressive lenses as one bill each. The share of money moving through UPI is a different question, and this data does not answer it.
2,883 orders carrying a recorded payment method, from one Indian optical business with two branches, 2022 to 22 September 2026. Counts of bills, not amounts.