Most restaurant software rollouts fail in the first two weeks because the team tries to do everything at once. The kitchen, the menu, the staff training, the printers, the payment gateway — all in parallel. By week 3 the chef is angry, the cashier is using paper again, and the owner is on the phone with support every morning.
We learned to sequence the work. Here is the rollout we run with every new outlet. It works.
Day 1 — setup
- 1Create your Indostra workspace. 10 minutes.
- 2Add the outlet, GST number, and bank account for UPI settlement.
- 3Print one test QR. Scan it. Confirm the (empty) menu loads on your phone.
- 4Stop. Go home. The setup is done.
Day 1 is for confirming the rails work. Menu work needs the chef present and unhurried. Save it.
Week 1 — menu
Block 4 hours with the chef. Import the menu — you can paste from a CSV or take photos of your existing menu and let our parser draft it. Review every item with the chef present: name, description, price, modifier groups, station tag, course number.
- Tag every item with one station (tandoor, curry line, bar, dessert, expo).
- Tag every item with a course (1 = starter, 2 = main, 3 = dessert).
- Add 'spice level' as a modifier on every Indian dish — guests ask anyway.
- Add great photos for your top-10 sellers. No photos for the rest.
Week 2 — kitchen
Install the printers. Connect each one to a station. Send 50 test KOTs and watch them land. Have the line cooks read 5 of them aloud — if anything is unclear, the layout needs work, not the cook.
Train the kitchen on what 'fire mains' means and who is responsible for tapping it. Most kitchens settle on the expo person doing it. Some have the senior server do it. Pick one and stick with it.
Soft launch
Pick a Tuesday. Run QR ordering for 10 tables only — mark the others 'Server only' for now. Have your manager shadow the floor with a notebook. Note every guest question, every staff hesitation, every printer hiccup.
Debrief at close. Fix the top 5 issues the next day. Run a second soft launch on Thursday with 20 tables. Repeat.
Go-live day
Saturday dinner. Full QR coverage. Pay-at-table on. Print 30 spare paper menus and put them at the host stand for guests who ask. You will not need most of them, but having them removes the panic.
Do not launch a new dish. Do not change a price. Do not introduce a new server. Hold every other variable constant so you can measure what the system did.
Day 30 retro
Pull the dashboard. Compare 30 days before to 30 days after on three metrics: average turn time, AOV per cover, and order error rate. Have the chef and the manager in the room. Discuss what to keep, what to change, and what to add next month — KDS for the bar? Modifier pruning? Loyalty integration?
The system is not done at day 30. It is just beginning to compound. Everything from here is iteration.
15 years in restaurant operations across 3 continents. Former GM of a 5-star hotel restaurant in Mumbai.