Hospitality (HoReCa)
14 released templates. Each passed a check on test data; it is switched on in the panel by the director, and a person decides on every proposal. Example results are illustrative: the names and companies in them are made up.
Table reservations
The problem it removes: Phone reservations get lost.
More detail
What it does: Reservations in chat and a reminder.
What a person decides: Confirmation.
For whom: Guest.
How to measure: Reservations without calls; no-shows.
Example result: “Got it: a table for 4 on Saturday at 19:00, by the window if possible. The host will confirm your reservation within 15 minutes here, and we'll remind you on the day. Is there an occasion — a birthday or a business meeting? We'll be ready.”
Menu questions
The problem it removes: Servers answer the same questions over and over.
More detail
What it does: Answers about ingredients and allergens from the menu.
What a person decides: The final decision on the outcome.
For whom: Guest.
How to measure: Questions resolved without a call.
Example result: “The Caesar has no nuts, but it does have egg, Parmesan and anchovies in the dressing. If you have an allergy, tell your server when ordering — the chef will check the ingredients and, if possible, make it without the item.”
Recipe cards and food cost
The problem it removes: Nobody knows what dishes actually cost.
More detail
What it does: Recipe cards and cost calculation.
What a person decides: Prices.
For whom: Chef, manager.
How to measure: Food cost accuracy.
Out-of-stock dishes
The problem it removes: Guests order dishes that are out.
More detail
What it does: An out-of-stock (86) list for the dining room and delivery.
What a person decides: The final decision on the outcome.
For whom: Staff member.
How to measure: Orders for out-of-stock dishes.
Example result: “Action: 86 the salmon and the signature honey cake — per kitchen stock at 16:00 there's salmon for 2 portions, while it's ordered 8–10 times in the evening; the honey cake is gone. Remove them from the delivery menu and tell servers to offer sea bass instead of salmon. Impact: guests aren't turned down after ordering. Risk: if a salmon delivery comes in tonight, someone may forget to lift the 86 — check at 19:00. Fallback: the chef lifts it after the delivery.”
Restaurant reviews
The problem it removes: Reviews on 2GIS go unanswered.
More detail
What it does: Replies and complaint follow-up.
What a person decides: Publishing.
For whom: Manager.
How to measure: Restaurant rating.
Example result: “Aizhan, thank you for writing. You're right: a 50-minute wait for a main course isn't our standard — we're sorry. We've found the cause and added kitchen staff on Fridays. We'd be glad if you gave us another chance — send us a DM and the host will reserve a table for you.”
Delivery status
The problem it removes: People call asking “where's my order”
More detail
What it does: Gives the customer the delivery status.
What a person decides: The final decision on the outcome.
For whom: Guest.
How to measure: Delivery status calls.
Example result: “Your order is being prepared — the courier leaves at 19:20 and will be with you in about 25 minutes. If anything changes, we'll let you know right away.”
Banquets and toi celebrations
The problem it removes: A toi or anniversary request is the biggest ticket, yet finding the answer on the hall, menu, and deposit takes half a day.
More detail
What it does: Answers banquet and toi requests: which hall fits the guest count, the per-person menu, deposit and prepayment, and what guests can bring — per the venue’s banquet offer.
What a person decides: Booking the hall, final menu, and price.
For whom: Guest.
How to measure: Banquets booked from chat inquiries.
Pre-orders and pickup
The problem it removes: A guest wants to grab food on the way home, but nobody picks up the phone at rush hour.
More detail
What it does: Takes takeout pre-orders: what was ordered, when it’ll be ready, how to pay, and where to pick up — per venue rules.
What a person decides: Kitchen confirmation of the order.
For whom: Guest.
How to measure: Takeout orders without phone calls.
Example result: “Got it: plov and two samsas to go, ready at 19:30. The kitchen will confirm within 10 minutes. Pay by Kaspi transfer or on site, and pick up at the counter to the right of the entrance.”
Daily menu and business lunch
The problem it removes: Every morning the manager types the daily menu into stories and the regulars’ chat by hand.
More detail
What it does: Turns the kitchen’s daily menu into a short post and a message for regulars: dishes, lunch price, hours — with no ingredient mistakes.
What a person decides: Posting.
For whom: Guest.
How to measure: Business lunch guests.
Kitchen waste
The problem it removes: Food gets written off every day, and nobody tracks how much money goes in the trash or why.
More detail
What it does: Shows from the waste log what gets written off most, for how much, and why — and what to change in purchasing and storage.
What a person decides: Changes to purchasing, storage, and the menu.
For whom: Head chef, manager.
How to measure: Monthly waste write-offs.
Nightly summary for the owner
The problem it removes: The owner hears how the evening went from scraps in the managers’ chat.
More detail
What it does: Builds the owner a daily summary from the POS export and shift log: revenue, guests, seating, voided dishes, complaints, and what to decide tomorrow.
What a person decides: Staff, menu, and pricing decisions.
For whom: Owner.
How to measure: Owner time reviewing the evening.
Kitchen logs for inspection
The problem it removes: Health inspectors show up unannounced, and the temperature and quality check logs have gaps.
More detail
What it does: Shows gaps and deviations in the week’s kitchen logs — fridge temperatures, quality checks, cleaning, staff health certificates — and what to fix first.
What a person decides: Corrections and clearing staff to work.
For whom: Head chef, manager.
How to measure: Gaps in kitchen logs.
Example result: “Situation: this week the temperature log for the meat prep fridge is blank on 3 of 7 days, and on October 9 it shows +9 °C. Key point: cook Yerlan’s health certificate expired October 5. Options: 1) send Yerlan to renew it today and find shift cover — fully compliant, but short-handed this weekend; 2) put off renewal until Monday — risk of a fine if inspected. Recommendation: 1. Today: assign someone to the temperature log and check the meat prep fridge.”
Office lunch catering
The problem it removes: Companies look for a place to order staff lunches, and the answer on menus and invoicing comes a day later.
More detail
What it does: Answers companies that want office lunch delivery: set meals and prices, minimum order, delivery time, invoiced payment — per venue terms.
What a person decides: Contract, corporate price, and delivery schedule.
For whom: Corporate customer.
How to measure: Companies on regular lunch orders.
Weekend bookings and staffing
The problem it removes: Friday’s dining room is packed with two servers — or the reverse, staff stand around idle.
More detail
What it does: Uses weekend reservations to show expected guests by hour, where the peak is, and whether there are enough servers and cooks on shift.
What a person decides: The shift schedule and calling in extra staff.
For whom: Manager.
How to measure: Weekend complaints about slow service.
Example result: “Action: call in a third server on Saturday from 18:00 to 23:00 — reservations show 64 guests between 19:00 and 21:00, but only two are scheduled, plus a 20-person banquet in the small room. Add a second cook on the hot line from 17:00. Impact: no delays at peak. Risk: if the banquet moves, the third server is extra — confirm with the client on Friday. Fallback: send the server home at 21:00 if the room is half empty.”
Basis: product documentation; the platform code. This page describes what exists in this build, not what is planned.