Rule Based Chatbot Example You Can Play With: A Bike Service Centre’s Bot, Every Rule and Message Included

Rule based chatbot example posts usually show a diagram and three sample messages. This one shows a complete bot: every button, every keyword rule, every message word for word, and the reasons behind each choice. It’s built for an illustrative two-wheeler service centre, and you can play with it on this page, either by tapping buttons or by typing like a customer. Then use the adaptation table to turn the same structure into a bot for your own business.

9button steps
6keyword rules
1fallback
4handover points

⚡ In short

  • A complete small-business bot needs only nine button steps, six keyword rules and one fallback.
  • Buttons do most of the work; keyword rules catch people who type instead of tapping.
  • Four points hand over to a person: booking confirmation, tracking, complaints and direct requests.
  • The same structure fits clinics, coaching centres, salons and repair shops with small changes.
Rule based chatbot example in LiveChatBot: AI Agent knowledge sources and a sample chat answered from them
How it looks in LiveChatBot: AI Agent knowledge sources and a sample chat answered from them.

The business

Who this bot is for, and what it must do

Speedy Bike Care is a made-up two-wheeler service centre in a mid-sized Indian city, with three mechanics and one front-desk person. It services scooters and motorcycles, offers pickup and drop, and gets most enquiries on WhatsApp and the phone. The owner adds website chat because the phone rings constantly with the same three questions: how much does a service cost, can you pick up my bike, and is my bike ready.

Before building anything, the owner writes down what the bot must achieve:

  1. Answer price and pickup questions instantly, with exact numbers.
  2. Turn booking requests into a phone number the front desk can call back.
  3. Never try to resolve complaints or tell someone their bike’s status. Those go to a person.
  4. Work after hours by collecting phone numbers for a morning call.

That’s it. Four goals, no clever conversation. Notice what’s missing: no attempt to diagnose problems (“my bike makes a noise”), no payments, no discounts. Those need a mechanic or the owner. A good rule based chatbot example is as much about what the bot doesn’t do as what it does.

Play with it

Try the bot: tap buttons or type like a customer

Try typing “kitna lagega service ka”, “my bike is not working after service” or “can you collect my scooter”.

Rule fired: –

The structure

The whole bot on one page

Five menu options from the greeting. Booking goes three steps deep; everything else is one step.

Greeting + 5 buttons
↓
Book a service
↓
Bike type
↓
Day
↓
Phone → team
Service prices
↓
Exact prices + book button
Pickup and drop
↓
Rules + book button
Track my bike
↓
Job card → person
Talk to a person
↓
Team or away form

Every message

All nine button steps, word for word

Copy and change the details for your own business.

RuleStepMessageButtons
R1Greeting / menuHi! I’m the Speedy Bike Care assistant. What do you need today?Book a service, Service prices, Pickup and drop, Track my bike, Talk to a person
R2Booking: bike typeGreat! Which bike is it?Scooter, Motorcycle up to 200cc, Above 200cc, Back
R3Booking: dayWhich day suits you?A weekday, Saturday, Back
R4Booking: collect phone → hand overPlease type your phone number. Our team confirms the slot within 30 minutes during working hours (9 am to 7 pm, Monday to Saturday).I’ve shared it, Start again
R5Booking: confirmationThanks! Your request is with our team. If it’s after 7 pm, we’ll call you first thing tomorrow.Start again
R6Prices (exact wording)General service: scooter ₹799, motorcycle up to 200cc ₹999, above 200cc from ₹1,799. Parts are extra and we quote before starting any work.Book a service, Pickup and drop, Back
R7Pickup and dropPickup and drop is free within 5 km and ₹149 up to 10 km. It isn’t available on Sundays.Book a service, Back
R8Tracking → hand overPlease share your job card number. A person will check your bike’s status and reply here.Talk to a person, Back
R9Hand over to a personConnecting you to our team now. Outside working hours, leave your number and we’ll call you back.Start again

Keyword rules

What happens when people type instead of tapping

About a third of visitors ignore buttons and type. These rules catch them, checked top to bottom.

OrderKeywords (whole words)Goes to
1problem, not working, damaged, complaint, refund, overchargedR0 · Complaint words (top priority) → Hand over to a person
2book, booking, appointment, slot, service myR2 → Booking: bike type
3price, cost, charges, rate, how much, kitnaR6 → Prices (exact wording)
4pickup, pick up, drop, collectR7 → Pickup and drop
5status, track, ready, job cardR8 → Tracking → hand over
6person, human, call, talk, managerR9 → Hand over to a person

Why complaint words come first: “my bike is not working after service” contains “service”, which would otherwise start a booking. The complaint rule catches “not working” first and hands over, which is what an unhappy customer needs. The matching traps behind this are explained in rule based chatbot rules that don’t misfire.

Design choices

Why the bot is built this way

Five buttons, not ten

Five options fit on a phone screen without scrolling. Anything rarer than these five topics goes through “Talk to a person”. Showing ten buttons makes people read instead of tap.

Prices in full, in one message

The price message includes all three bike classes and the “parts extra” line. Splitting prices across several steps makes people tap three times to learn one number, and hiding “parts extra” creates arguments at the counter.

Booking collects only three things

Bike type, preferred day and phone number. Not the registration number, not the model, not the problem description. The front desk asks the rest on the call. Every extra question loses a few people before they finish.

Tracking always goes to a person

The bot can’t see the workshop. Rather than pretending, it asks for the job card number and hands over. Honest handovers build more trust than vague automated answers. See conversation bot handover for patterns.

A “Back” button on every step

People tap the wrong option. Without “Back” they have to restart the chat, and many simply leave.

The fallback offers a way out

“Sorry, I didn’t catch that. Pick an option below, or tap ‘Talk to a person’.” It never guesses, and it always shows the buttons again.

After hours

The same bot at 10 pm

Only the handover messages change. Opening hours decide which version is shown.

Booking confirmation
🚩 Working hoursOur team will confirm within 30 minutes.
✅ After hoursThanks! We’re closed now. We’ll call you between 9 and 10 am tomorrow to confirm your slot.
Talk to a person
🚩 Working hoursConnecting you to our team now.
✅ After hoursOur team is offline until 9 am. Leave your number and we’ll call you first thing.
Tracking
🚩 Working hoursA person will check your bike’s status and reply here.
✅ After hoursThe workshop is closed. Leave your job card number and phone, and we’ll update you by 10 am.

In LiveChatBot, opening hours switch the chat to the away form automatically, and messages left after hours can become tickets the team sees in the morning.

Results to expect

What the owner checks after 30 days

A bot like this is judged on a few plain numbers, all visible in the inbox.

Share of price and pickup questions answered without a person. These should almost all be handled by the bot. If people still ask the front desk about prices after reading the bot’s answer, the message isn’t clear enough.

Booking requests with a phone number. Count how many people who start “Book a service” reach the phone step. If many drop off at “Which day suits you?”, consider removing that step and asking on the call instead.

Fallback rate. How often does nobody’s keyword match? Read those chats every week. Each repeated fallback topic is a new keyword or a new button.

Complaints caught early. Check that every unhappy message reached a person quickly. One complaint answered with a booking prompt does more damage than a hundred good automated answers.

The owner keeps a short weekly note: what fell through to the fallback, which keywords to add, which message to rewrite. Most bots reach a stable state within a month of these small weekly fixes.

Adapt it

The same structure for other businesses

Keep the shape: greeting menu, one booking path, exact-price answer, a status question that goes to a person, and a human button.

BusinessBooking path asksExact-answer topicsAlways to a person
ClinicDoctor, day, phoneFees, timings, locationSymptoms, reports, complaints
Coaching centreCourse, batch, phoneFees, batch times, demo classRefunds, results, scholarships
SalonService, day, phonePrice list, timings, offersBridal packages, complaints
Mobile repairPhone model, issue type, phoneScreen and battery prices, warrantyRepair status, data recovery
BakeryCake size, date, phoneFlavours, prices, delivery areaCustom designs, allergies

For a deeper walk-through of one of these, see our coaching institute bot built step by step or the bakery assistant built start to finish.

Build it yourself

From this example to your own bot in an afternoon

Start by copying the structure, not the words. Write your five greeting buttons from your five most common questions. Write one exact-answer message for each simple topic, with real numbers. Decide which three details your booking path needs. List your complaint words and your “talk to a person” words. Then write the fallback.

In LiveChatBot you build the buttons with Chat Flows, store exact answers as saved replies so the free keyword bot can use them, and route handovers to your team’s inbox and Android app. The AI Agent can sit on top for open questions, while this rule based chatbot example structure keeps the important paths predictable. The free plan covers two websites with 100 visitors per website each month. Start on livechatbot.in.

Frequently asked questions

What is a good rule based chatbot example for a small business?

One with a short greeting menu, a booking path that collects a phone number, exact answers for prices and policies, and clear handovers for complaints and status questions, like the bike service bot on this page.

How many buttons should the first message have?

Around five. Enough to cover your most common topics, few enough to fit on a phone screen.

Do I still need keyword rules if I use buttons?

Yes. Many visitors type instead of tapping, so keyword rules catch them and send them to the right step.

What should the bot do when nothing matches?

Apologise briefly, show the menu buttons again, and offer a person. Never guess.

Can I copy this rule based chatbot example for my own business?

Yes. Keep the structure, then replace the topics, prices, booking questions and handover messages with your own. Test it by typing like a real customer before going live, and read the fallback chats every week for the first month.

Can I add AI later?

Yes. Keep the buttons and rules for booking, prices and handovers, and let AI answer open questions from your knowledge.

Keep reading

Related guides

📘

Rule Based Chatbot Rules That Don’t Misfire: Keywords, Order, Negatives and a Free Rule Tester

Read the guide →

🧭

Robot Chatbot Confusion: Physical Robots, Software Bots and RPA Explained for Business Owners

Read the guide →

🛠️

Questions for Chatbot Conversations: What Your Bot Should Ask Visitors, and When

Read the guide →

Build your own version of this bot

Copy the structure, swap in your topics and numbers, and test it like a customer. Start free on livechatbot.in.