An illustrative walkthrough: taking a 30-cover cafe from paper to QRated in a weekend
A worked example on our own demo tenant, Sunrise Cafe: how the migration weekend goes, what tends to break, and what to measure in the first 30 days. Sample data, not a customer story.
- walkthrough
- getting-started
- bengaluru
This is an illustrative walkthrough on our own demo tenant, Sunrise Cafe, the menu we use to show the product. It is not a customer story. Every figure below is sample data, chosen to show what to measure, not what anyone achieved. Our one live customer is Cafe 2o6, and we have not asked them for numbers.
Picture a 30-cover cafe in Bengaluru. Two co-owners, one on the floor and one in the kitchen, six staff, one senior cook. They have run on a printed menu and chits on a clip since opening. This is how a weekend switch to QRated goes, step by step, using the Sunrise Cafe demo menu as the example.
Why paper stops working
Three things push a cafe like this off paper.
1. Modifier confusion. Sugar level on chai, milk type, spice level, half portions. Servers write shorthand on the chit, the kitchen misreads it, the dish comes back. Every return is a remake and a comp.
2. Price changes. A menu that changes monthly means a reprint, or a taped-over price, or a server explaining that the printed number is wrong. Most cafes run with at least one stale price on the table.
3. No item-level numbers. The cook has a feel for what sells. The owners have a spreadsheet. Nobody has hard counts per item, so the menu never gets pruned.
The migration weekend
Friday evening (about 3 hours)
- Sign up. Every new account gets 14 days of Starter, no card.
- Pick a theme. Three are free; the demo uses Bakery Warm.
- Set the brand colour from the signboard.
- Add the categories: Chai and Coffee, Snacks, Bakes, Cold Drinks.
Saturday morning (about 4 hours)
- Photograph every dish on one plate, in daylight, against the counter. For 40 items this is a morning.
- Add the items. The spreadsheet import gets a typed-up menu live in about ten minutes; typing a full menu in by hand is an afternoon.
- Set up sizes and add-ons (Starter and up): Cutting or Full on chai, sugar levels, oat milk at a supplement.
- Turn on the help button: Call waiter, Refill water, Bring the bill.
Saturday evening (about 1 hour)
- Download the table tent PDFs (A5 landscape, one per page) and the QR stickers. Nothing physical ships; you print them or send the PDF to the printer down the street.
Sunday morning (about 1 hour)
- Put a tent or sticker on each table.
- Scan with three different phones and confirm the menu opens in the browser with no app.
- Mount the kitchen tablet near the cook station. Any device on Chrome 110+, Safari 16+ or Edge from 2019 or newer works; no printer.
- Brief the cook: from lunch, no more chits, read the screen.
Sunday lunch
The first scan comes in. Orders show on the Orders page on every plan; on the Starter trial the live kitchen board is on too, with table number, age and every customisation spelled out. Tickets turn amber when they wait too long.
What tends to go wrong on day one
- A cook taps Ready on the wrong ticket. One tap moves a ticket along, so a wrong tap is easy. The fix is a habit, not a setting: read the table number before tapping.
- A guest pays and the ticket is slow to appear. Refresh the kitchen screen; it checks Cashfree for any order still waiting on payment. The FAQ has the full answer.
- The cafe wifi is slow. Guests on mobile data keep getting the menu, but the kitchen tablet needs a working connection. Sort the wifi before the stickers go on tables.
The numbers worth watching in the first 30 days
These are the columns to fill in from your own dashboard. The figures here are sample data to show the shape of the table, not results.
| Metric | Before (paper, your estimate) | After 30 days (from Analytics) |
|---|---|---|
| Average ticket size | sample data | sample data |
| Items per order | sample data | sample data |
| Kitchen comps and remakes | sample data | sample data |
| Median time to table | unknown on paper | sample data |
| Returning guests | unknown on paper | sample data |
| Menu reprint cost | sample data | ₹0 |
Analytics and the returning-guest count are on Starter and Pro. On Free, the Orders page and a CSV export give you the order-level rows to compute the first three yourself.
What to expect from staff
The senior cook usually wants more data, not less. Once the board shows which ticket is due first, the next question is which station is slow.
The longest-serving waiter takes the longest. A chit-on-clip habit of ten years does not change in a day. A five-minute one-to-one beats a group briefing.
What we would tell any operator about to migrate
- Do not launch the full menu. Start with the 20 best sellers and add the rest in week two.
- Fix the wifi before the QR codes go on tables. A slow menu is worse than a paper one.
- Brief every staff member individually.
The honest economics
What QRated costs: Free is ₹0 with no card and no expiry. Starter is ₹7,500 + 18% GST = ₹8,850 a year (about ₹24 a day) and adds the live kitchen board, sizes and add-ons, analytics and coupons. Pro is ₹10,000 + 18% GST = ₹11,800 a year and adds numbered GST invoices and the AI menu tools. Online payments carry a 2.5 to 3% platform fee plus GST on the fee; counter orders carry nothing. One outlet per account.
What it replaces: menu reprints, and the remakes that come from misread chits. How much that is worth depends entirely on your menu and your volume, which is why the table above has no numbers in it.
What this walkthrough is not
It is one shape of cafe: single outlet, modifier-heavy, owner-operated. A 200-cover dining hall, a delivery-only kitchen or a bar would have a different curve, and we have not run this with any of them. When a customer agrees to share real numbers, we will publish them under their name.