Restaurant Chatbots: What They Handle, How to Set One Up

A restaurant chatbot answers guest questions instantly — menu items, allergen information, hours, reservation requests, catering details — directly on your website, around the clock. It handles the repetitive informational load so your staff can focus on the guests already in front of them.

Multiply the calls your restaurant receives daily asking about hours, parking, private dining, allergen information, and whether you take reservations. Each one pulls someone away from serving the guests who are already in your restaurant. A chatbot trained on your menu, hours, and policies handles those questions without interrupting your team.

What a restaurant chatbot handles

The table below maps the most common restaurant use cases to what the chatbot manages and the direct benefit to your operation.

What visitors ask your restaurant

Before building a chatbot, it helps to map out the questions your restaurant actually receives. Most fall into a predictable set of categories.

A chatbot trained on your website, menu, and FAQ content can answer every one of these accurately and immediately.

Use cases that save your team real time

Menu Q&A

Your menu is your most-visited page, and it generates the most questions. A chatbot trained on your menu PDF or menu page can answer specific questions like:

• "What appetizers do you have under $15?"
• "Is the Caesar salad dressing dairy-free?"
• "What comes with the grilled salmon?"
• "Do you have a prix fixe menu?"
• "What are tonight's specials?"

For specials and rotating items, add a text source that you update weekly with current specials. This keeps the chatbot accurate without retraining the entire menu.

The allergen and dietary question is particularly valuable. A visitor with a severe allergy who cannot get a quick answer about ingredients will choose a different restaurant. A chatbot that responds immediately with "Our pad Thai is prepared with peanuts and is not suitable for guests with peanut allergies. However, our rice noodle stir-fry can be prepared peanut-free. Please inform your server about your allergy when you arrive" builds confidence and keeps the reservation.

Important caveat: Always include a disclaimer that guests should confirm allergen information with their server, as recipes and preparations can change. Your chatbot should say something like: "Please inform your server about any allergies when you arrive so we can accommodate your needs safely."

Hours and holiday hours

This sounds trivial, but it generates a disproportionate number of calls and website visits. A chatbot handles it effortlessly:

• Regular hours for each day of the week
• Holiday hours and closures
• Happy hour times
• Brunch, lunch, and dinner service windows
• Kitchen close time versus bar close time
• Seasonal schedule changes

The real value here is holiday hours. Every year, restaurants get flooded with calls asking "Are you open on Thanksgiving?" and "What are your Christmas Eve hours?" A chatbot with updated holiday information saves your team from answering the same question dozens of times.

Add a text source at the start of each holiday season with your holiday schedule. This takes five minutes and saves hours of phone time.

Reservation requests

If your restaurant takes reservations, a chatbot can streamline the request process. This is not about replacing OpenTable or Resy. It is about capturing reservation requests from visitors who might otherwise leave your site without taking action.

Here is how it works with a lead capture action:

1. Visitor asks about reservations
2. Chatbot confirms you take reservations and asks for details
3. Lead capture form collects: name, phone number, party size, preferred date and time
4. Your host receives the request and confirms or suggests alternatives

The lead capture form fields for a restaurant reservation request:

This works alongside your existing reservation system. The chatbot captures the request; your staff confirms it through your normal process. The same lead capture mechanic works across hospitality contexts — see how hotels use chatbots for room bookings and guest services for a comparable approach.

Catering inquiries

Catering inquiries are high-value but time-consuming. A prospective client wants to know your minimum order size, price ranges, menu options, delivery radius, and setup services. These conversations often require multiple back-and-forth emails before anyone commits.

A chatbot trained on your catering menu and policies handles the initial information exchange:

• Catering menu options and pricing tiers
• Minimum order sizes (e.g., "Our catering minimum is 10 guests")
• Delivery area and fees
• Setup and service options
• Lead time requirements ("We require 48 hours advance notice for catering orders")
• Payment terms and deposit information

Once the visitor has the information they need, the chatbot captures their contact details and event specifics through a lead capture action, giving your catering manager a qualified lead with context instead of a cold inquiry.

Private events and group dining

Restaurants with private dining rooms or event spaces receive regular inquiries about capacity, minimum spend requirements, menu options, and availability. Train your chatbot on:

• Room capacities and configurations
• Minimum spend or buyout requirements
• Private dining menu options
• Audio/visual equipment available
• Deposit and cancellation policies
• Photos or virtual tour links

The chatbot answers preliminary questions and collects event details (date, guest count, occasion, budget range) through lead capture, giving your events coordinator everything they need to follow up with a proposal. If you run a business where appointments and bookings are central, the same chatbot flow applies — the guide on chatbots for appointment booking covers the pattern in detail.

Setting up your restaurant chatbot

Step 1: Gather your content

Before creating the chatbot, collect the information you want it to know:

• Your website URL (for crawling hours, location, about, menu pages)
• Your menu in PDF format (especially useful if your menu page is image-based)
• A text list of current specials (if they rotate)
• Your reservation and private dining policies
• Catering menu and policies (if applicable)
• Your top 20 most frequently asked questions and their answers

Step 2: Create your chatbot

Sign up for Agentkit and create a new chatbot. Name it after your restaurant: "Bella Vista Assistant" or "The Oak Table Bot." Training works via website crawl and document upload — paste your URL, upload your menu PDF, and Agentkit indexes both.

Step 3: Add training sources

Website crawl: Enter your restaurant's website URL. Agentkit will crawl your pages and index the content. This picks up hours, location, menu items (if text-based), and general information.

Menu PDF upload: Upload your full menu as a PDF. This is critical if your website uses images for menu items rather than text. The PDF gives the chatbot access to every dish, price, and description.

Q&A pairs: Add your most common questions. Focus on the ones your staff answers by phone multiple times per week:

• "Do you have outdoor seating?" / "Yes, we have a covered patio that seats 30 guests. It is available on a first-come, first-served basis."
• "Is there parking?" / "We have a parking lot behind the building with 40 spaces. Street parking is also available on Main Street."
• "Do you accommodate large groups?" / "We can seat groups of up to 20 in our main dining room. For groups larger than 12, we recommend making a reservation. We also have a private dining room for events."

Text sources: Add current specials, holiday hours, and any frequently changing information.

Step 4: Configure system instructions

Write instructions that set the chatbot's tone and boundaries:

"You are the virtual assistant for [Restaurant Name]. You help guests with menu questions, hours, location, reservations, catering, and private events. Use a friendly, welcoming tone. Answer questions using only the information in your training data. For allergen questions, always remind guests to confirm with their server. If you do not know the answer to a question, suggest the guest call us at [phone number]. Do not make up menu items, prices, or policies."

Step 5: Set up lead capture for reservations

Configure a lead capture action for reservation requests. Set the trigger so the chatbot offers it naturally when someone asks about booking a table. The form should collect name, phone number, party size, preferred date, and preferred time. Lead capture is available on all Agentkit plans, including the free tier.

Step 6: Add suggested messages

Help visitors start the conversation with suggested messages:

• "What are your hours today?"
• "Show me the dinner menu"
• "Do you have gluten-free options?"
• "I would like to make a reservation"
• "Tell me about private events"
• "Do you offer catering?"

These prompts guide visitors toward the chatbot's strongest topics and reduce the chance of off-topic questions.

Step 7: Customize the widget

Match the chatbot's appearance to your restaurant's branding. Set colors that complement your website, add your logo, and write a welcome message:

"Welcome to [Restaurant Name]! I can help with our menu, hours, reservations, and more. What can I help you with?"

Step 8: Embed on your website

Add the script tag to your website. If your site runs on WordPress, use the WordPress plugin. The widget loads asynchronously so it does not affect your page speed. The chatbot appears as a chat bubble, typically in the bottom-right corner of the screen.

Step 9: Test with real questions

Before going live, test with 30-40 questions your restaurant actually receives:

• Verify menu items and prices are accurate
• Confirm hours are correct, including any seasonal variations
• Test allergen questions and check that the disclaimer appears
• Test the reservation lead capture flow end to end
• Ask about catering and private events
• Try questions in other languages — Agentkit auto-detects the visitor's language and responds in kind, supporting 95+ languages

Multi-location considerations

Restaurant groups with multiple locations face a decision: one chatbot or many?

One chatbot for all locations works if:

• All locations share the same menu
• Hours are consistent across locations
• You want a single embed across your corporate website

The chatbot needs clear training data that distinguishes between locations. Include location names in your content: "Our downtown location is open until 11 PM on weekends. Our midtown location closes at 10 PM."

Separate chatbots per location works better when:

• Menus differ by location
• Hours vary significantly
• Each location has its own website or landing page
• Catering and private event offerings differ

With separate chatbots, each location gets tailored responses without the risk of cross-contamination (telling a visitor at Location A about Location B's happy hour).

A hybrid approach: Use one chatbot on your corporate website with suggested messages like "Which location are you visiting?" and separate chatbots on individual location pages.

The multi-location challenge is common across hospitality. The hotel chatbot guide covers similar tradeoffs for properties with multiple room types and seasonal policies.

Keeping information current

Restaurant information changes frequently: seasonal menus, holiday hours, special events, price adjustments. An outdated chatbot erodes trust.

Weekly updates: Review and update specials, seasonal items, and any changed policies. This takes five minutes if you maintain a simple text source for rotating content.

Seasonal updates: At the start of each season, update your menu PDF if the menu changes, and update any seasonal hours.

Holiday updates: Before each major holiday period, add a text source with your holiday schedule. Remove it after the holidays pass.

Auto-retrain: On Agentkit's Standard plan and above, auto-retrain re-crawls your website automatically. If your website stays current, your chatbot stays current without manual intervention.

What a chatbot does not replace

A chatbot does not replace the warmth of a great host, the expertise of a knowledgeable server, or the personal touch that turns first-time visitors into regulars. It handles the informational questions that do not require that personal touch, freeing your team to focus on the guests who are already in your restaurant.

Think of it as a knowledgeable assistant who never takes a break, never puts anyone on hold, and never gets frustrated answering "What time do you close?" for the fortieth time today.

For non-restaurant hospitality contexts, the same principles apply — see how customer support chatbots handle high-volume repetitive inquiries across service industries.

Frequently Asked Questions

Do restaurant chatbots work in multiple languages?

Yes. Agentkit automatically detects the language a visitor writes in and responds in the same language. This works across 95+ languages without any configuration — helpful for restaurants in tourist areas or neighborhoods with diverse communities.

Can a chatbot take actual reservations, or just collect requests?

A chatbot can collect reservation requests through a lead capture form — name, party size, preferred date and time, phone number — and route those to your team for confirmation. It does not directly integrate with OpenTable or Resy to book in real time, but it captures the request 24/7 so nothing falls through the cracks. For a deeper look at the booking flow, see chatbots for appointment booking.

How do I keep the chatbot accurate when my menu changes?

Upload a new menu PDF whenever your menu changes, or maintain a text source with your current specials that you update weekly. On the Standard plan ($119.99/month), auto-retrain re-crawls your website on a schedule so changes you make to your live site are picked up automatically.

What does a restaurant chatbot cost?

Agentkit offers a free plan ($0) to start, a Hobby plan at $29.99/month, and a Standard plan at $119.99/month. All plans include lead capture for reservation and catering inquiries. Higher plans add more chatbots, more training sources, and auto-retrain.

What happens when the chatbot does not know the answer?

You configure a fallback response in the system instructions — typically something like "I'm not sure about that. Please call us at [phone number] or email [address] and our team will help." This keeps the guest experience clean even when a question falls outside the chatbot's training data.

Getting started

Your restaurant's website already has most of the information a chatbot needs. Your menu is there. Your hours are there. Your location is there. A chatbot makes all of that information conversational and accessible.

Start with your menu PDF, your hours, and your top 20 FAQs. Add a reservation request form through lead capture. Embed the chatbot on your homepage and menu page. The setup takes less than an hour, and it starts answering questions immediately.

Build your chatbot for free →

No credit card required.