How a restaurant uses inTusell: daily operations guide

In the main article, you trained the inTusell AI: you uploaded your menu and knowledge base, set up working hours and reservation lengths, and configured the persona and response rules. Training was a one-time setup. Now the real work begins: how does a restaurant or café use inTusell, that is, how does restaurant inTusell usage work in daily operations? This article shows, step by step, what the assistant you trained actually does in the field over a working day, from the incoming DM to the table reservation, from a takeaway order to a reminder, to reducing no-shows and handing off to a human.

This article is the second part of the restaurant section of our sector-based "how to train your AI" series. The first part, how to train the inTusell AI, covered the setup; this one covers turning that setup into daily operations. We’ve written the same series for the tour agency and clinic sectors too.

Quick answer

A day in the flow

A restaurant that has set up inTusell has a typical day that looks very different from before. In the past, the owner or the floor manager would, at every spare moment between service, sift one by one through unread DMs, unanswered WhatsApp messages, and "are you open today?" comments. Now the messages that come in overnight and during service are already answered, reservation requests are collected (with the AI booking tools on, the reservations are opened too), and reminders are scheduled. The backbone of a day works like this:

The team is no longer busy sifting through messages, but instead deals with conversations that genuinely require a decision: a private event, a large group, a complaint. They no longer have to stand by the phone keeping up with DMs during service.

Who is it for?

This usage model makes the most sense especially for these businesses:

• Restaurants that receive heavy reservation demand via Instagram and WhatsApp: those who can’t keep up between service with dozens of "is there a table?" messages a day.
• Venues that suffer from no-shows: businesses where tables left empty by reservations that don’t turn up turn directly into lost revenue, especially on weekends.
• Cafés that lose after-hours and in-service demand: those losing guests because questions that come in during the evening rush or while closed are answered late.
• Businesses with high takeaway / pickup volume: those who have to take orders constantly via WhatsApp and note them down one by one.

If you have a volume of demand that a single person can’t keep up with between service, inTusell works like an experienced floor manager who handles that demand.

Incoming demand: the guest asks, the AI gives the right answer

The flow starts when the guest sends a message. Imagine someone writing on Instagram DM, "We’re coming as 6 on Saturday evening, is there a spot on the garden side?" The AI reads this message, scans the working hours, seating information such as garden and indoor areas, and venue rules in your knowledge base, suggests the option that fits the guest’s need, and steers toward a reservation.

There are two important points. First: the AI does not invent something that isn’t on the menu or in the knowledge base. To "Do you have gluten-free pizza?" it doesn’t guess if it isn’t sure; it says what it doesn’t know and hands off to the live team or the business. This is the most important trust message: rather than giving the guest wrong information and creating disappointment at the table, it’s safer to hold the answer back. Second: whatever language the guest writes in, the AI replies in that language; for a café in a tourist area this is critical, and no extra setting is needed.

If during training you added the response rule "before the reservation, clarify the party size, date, and time," the AI asks those three things first; the offer isn’t random but based on the real request. We cover separately how all these channels come together in a single inbox in the Instagram and WhatsApp automation article.

Table reservation: slot, capacity, and conflict prevention

When the guest asks for a time, the AI doesn’t pull a time out of thin air. It first fills in what is missing: date, time, party size and requests such as "window seat" or "birthday". What happens next depends on your account. If the AI booking tools are on, it looks at what you defined during training: your working hours, the reservation length (for example "Dinner 90 min") and whether that time is taken. In that case it books the free slot itself and writes the party size and requests into the reservation note; if the time is taken, it suggests the nearest free times. If the tools are not on, it passes the request to your team with those details, and you confirm the time or the guest picks a free slot on your online booking page. To see which applies, ask for a reservation in AI Test.

Here the most critical safeguard is conflict prevention: a booked time cannot take a second reservation. Before creating a reservation, inTusell checks your working hours and existing reservations; if a booked time is requested, no reservation is opened. Cancelled or no-show (missed) reservations free the time again in this check. If you take reservations for several tables at the same time, you define that service as a group session with capacity (for example "Saturday 20:00, capacity 10"); bookings count against the remaining capacity, and once it is full no new booking goes into that session. Capacity counts reservations, not people; inTusell has no table plan or table assignment, which you handle on the floor. So the nightmare that venues working with a manual reservation book experience most often (writing down more reservations for a time than you opened) disappears.

Every reservation has a status, and its lifecycle is clear:

scheduled (planned)
  → confirmed (confirmed)
  → attended (turned up / seated)
or
  → cancelled (cancelled)
  → missed (didn't turn up / no-show)

You’ll find detailed instructions on how to set up the reservation flow, reservation lengths, and calendar sync in the restaurant reservation automation article.

Takeaway and pickup order routing

Alongside reservations, the second main flow is the takeaway / pickup order. Imagine a guest writing on WhatsApp, "2 lahmacun, 1 ayran, takeaway." The AI reads the menu from the knowledge base, clarifies the order items and quantities, asks for the missing information (address, pickup time, note). If your menu is set up as products in the panel and Payment tracking is on, it confirms the items and amount with the guest, opens the order and sends the guest a tracking link to see it.

Here the boundary is clear: the AI never invents a price or stock. It quotes the price of an item defined in the menu from the knowledge base; if it isn’t defined, or a real-time stock question like "is it available today?" comes in, it doesn’t guess. It routes to the live team, or holds off by saying "let me confirm this with the kitchen." Once the order is opened, payment and delivery stay in your flow: the order lands in Orders in the panel, and the team sees it at a glance. If you take bank transfers, the AI sends the amount, your IBAN and a reference code in one message; your team approves the receipt.

So on the takeaway side, inTusell works like a front-line step that takes the order and records it neatly; carrying the result to the kitchen or the courier stays under the business’s control.

Reminders and reducing no-shows

The most expensive thing in a restaurant is a reservation that doesn’t turn up: an empty table is both lost revenue for that evening and a missed chance for another group you could have seated at that time, especially on a full Saturday night. On paid plans inTusell reduces this with automatic reminders.

For every reservation, two reminders are scheduled:

• 1 day in advance: a message along the lines of "You have a reservation tomorrow at [date time]" that asks for confirmation: "1 Yes · 2 Postpone · 3 Cancel".
• 2 hours in advance: a short reminder along the lines of "You have a reservation today at [time]: 2 hours to go".

The reminder goes through the channel the guest wrote on; if that channel is unreachable and an SMS provider is set up, SMS kicks in, and if that fails too, email. The message contains a cancel/reschedule link, so the guest can let you know in advance that they can’t make it and the table doesn’t go to waste. The guest manages their own reservation from the /manage-appointment/{token} page in the link; there’s no need to call your team. If you want to turn off the reminder for a specific reservation (reminders_enabled), you can do that per reservation too.

If the guest still does not show up, you mark the reservation with the "No-show" button in the Agenda or Day view; a reservation still not marked "Attended" about 2 hours after it ends also counts as missed, and the conversation moves to the "needs follow-up" stage. If you mark attendance in the panel (at least one "Attended" in the last 30 days), inTusell sends the guest who did not show a rescheduling message with up to 3 numbered options from your real open times and a reservation management link; if you do not mark attendance, it does not send it, because "missed" is then only a guess and telling someone who did come "we missed you" is worse than not writing at all. The guest can move the missed reservation to a new time from their own link within 7 days.

Instagram comments: a channel with its own switch

In a restaurant, the flow often starts with a comment under a post: "What are the prices?", "Do you take reservations?", "Are you open today?". In inTusell, WhatsApp, Instagram DM and Instagram comment automation are live today. DM replies are already on once Instagram is connected; comments are controlled by the "Instagram comments" switch on the Instagram card, which is on by default; you can turn it off.

The flow works like this: the AI gives one short public reply to suitable comments ("Of course, send us a DM for the details"). When the guest sends a DM, the conversation continues there and the AI replies, so topics like price and reservations get clarified in the DM. Replying to a comment by private message (Reply with DM) will open after Meta approval; today the AI writes a public reply and invites the customer to send a DM. The AI does not reply to spam comments with links; you hide or delete them from the panel. If you connected Instagram before September 24, 2026, the panel shows a "Reconnect Instagram to manage comments" warning; until you reconnect once, no comment gets an AI reply. Channel setup is detailed in the Instagram and WhatsApp automation article.

Following up a quiet guest: proactive follow-up and CRM

In a restaurant, most reservations are lost with guests who say "let me check and get back to you" and disappear. inTusell doesn’t forget these guests. A guest who got the information and then went quiet gets at most 3 follow-up messages written by the AI; they go out only within business hours on WhatsApp, Instagram or Telegram and stop when the guest replies, books, or asks not to be contacted. Every conversation is logged to the CRM, so the guest’s history, which date/event they were interested in, and what stage they’re at stay on record.

This way, when the team starts a new day, it’s clear which reservation is confirmed and which request is waiting on follow-up. There’s no need to keep a manual reminder list; a returning guest is recognized along with their past preference.

Handing off to a human: working modes

You decide how much the AI does on its own. Account-wide there are three modes (My business › Instructions › AI Settings › Advanced AI Settings):

In every mode the AI hands a conversation it is unsure about or one that needs a person, such as a request for a private event for 30 or a complaint, to your team. The handoff lands on the Decisions screen with the full conversation history; the guest doesn’t have to start over. If you want a single conversation handled differently, switch it to Approval, Shadow or Paused.

What it isn’t

To put inTusell in the right category, let’s clarify what it isn’t:

• It isn’t a simple chatbot. inTusell is an AI sales assistant and CRM that answers messages on WhatsApp, Instagram and web chat, books appointments and opens orders, and follows each customer through the sales pipeline.
• It isn’t a POS / check system. It doesn’t open tables or close checks, doesn’t drive a kitchen display; it’s the reservation, order-taking, and guest-communication layer.
• It isn’t a food marketplace / courier platform. It has no inventory of its own; it uses your menu and your rules, and leaves delivery to your flow.
• It isn’t a restaurant-specific priced product. Plans differ only by messages, voice minutes and storage, and don’t change by sector; channel count is never charged. The reservation and order module can be enabled on any plan.

So you don’t start with the wrong expectations, let these boundaries be clear from the outset. For product and pricing details, you can check the solutions and pricing pages.

Frequently asked questions

Does the AI open the reservation on its own?

If the AI booking tools are on in your account, yes: it suggests a free time based on your working hours and booked times, opens the reservation and writes the party size into the note. If not, it passes the request to your team and you confirm the time, or the guest picks a free slot on your online booking page. Double-booking the same time is prevented by the system; for several tables at the same time you define a session with capacity. You assign tables on the floor. For a large group or a private event it hands the conversation to your team.

How are no-shows (reservations that don’t turn up) reduced?

For every reservation, an automatic reminder goes out 1 day and 2 hours in advance; the guest can cancel or reschedule from the link in the message. A reservation still not marked "Attended" about 2 hours after it ends counts as missed, pending reminders are cancelled and the conversation moves to the "needs follow-up" stage. If you mark attendance in the panel (at least one reservation marked "Attended" in the last 30 days), the guest who did not show gets a rescheduling message with up to 3 options from your real open times and a reservation management link; if you do not mark attendance, this message is not sent, because "missed" would only be a guess.

Does the AI take takeaway / pickup orders?

Yes. If your menu is set up as products in the panel and Payment tracking is on, the AI confirms the items, quantities and amount with the guest, opens the order and sends the guest an order tracking link. Payment and delivery stay in your own flow. The AI never invents a price; it says an item not on the menu isn’t available.

Which channel does the reminder go through?

First through the channel the guest wrote on (WhatsApp, Instagram, Telegram); if that’s unreachable and an SMS provider is set up, then SMS, and if that fails, email. The guest doesn’t need to install a new app. Reminders work on paid plans; on the Free plan automations are off.

Can the guest be connected to a human?

Yes. When the AI is unsure or a conversation needs a person, it hands off to your team; the handoff lands on the Decisions page with the full conversation history. You can take over any conversation at any time, or switch just that conversation to Approval, Shadow or Paused.

Do reservations land in Google Calendar?

Yes. When a reservation is created, it’s copied to Google Calendar under the guest’s name; reschedules and cancellations are reflected too. The sync is one-way (inTusell → Calendar); the team sees the reservations in their own calendar, but an event added in Google does not block the time in inTusell.

Next step

If you haven’t trained your assistant yet, start first with the how to train the inTusell AI article: the foundation of this operation is set up there. If you want to go deeper on the channel side, you can move on to the Instagram and WhatsApp automation article, and for the details of the reservation engine, the restaurant reservation automation article; you can see the restaurant solution page and the whole series on the all articles page. You can review the price on the pricing page.

To see live how it would work in your own venue, request a demo or write directly to hi@intusell.com. In a short session, we open your inbox together and test the first reservation in the system.