For most Indian customers, WhatsApp is the default way to talk to a business. They message a shop to check stock, a clinic to book an appointment, a coaching institute about fees and a supplier about an order. That convenience creates a problem for growing businesses: the same questions arrive all day, often late at night, and replies slip as the team gets busier.
A WhatsApp chatbot solves this by answering routine questions instantly and passing real conversations to your staff. This guide covers what SMEs need to know before building one — the technology choices, costs, rules, use cases and a practical launch plan.
What Is a WhatsApp Chatbot for Business?
A WhatsApp chatbot is software connected to your business WhatsApp number that reads incoming messages and replies automatically. Simple bots follow menus ("Reply 1 for prices"). Modern AI chatbots understand free-text questions in natural language — including Hindi, Marathi and Hinglish — and answer from your own information: product catalogue, price list, policies and FAQs. The best ones also know when to stop and hand the conversation to a person.
WhatsApp Business App vs WhatsApp Business Platform
This is the first decision, and it confuses many businesses.
- WhatsApp Business app: free, runs on a phone, good for small volumes. It offers greeting messages, quick replies and a catalogue, but it is not designed for AI chatbots or deep integration with your systems.
- WhatsApp Business Platform (often called the WhatsApp Business API): Meta's official platform for businesses that need automation, multiple agents, integrations and higher volume. AI chatbots run here, usually through an approved provider.
Avoid unofficial tools that automate a regular WhatsApp account. They break WhatsApp's terms and put your number at risk of being banned — and losing the number your customers know can be far more expensive than doing it properly.
How Much Does a WhatsApp Chatbot Cost?
There are three cost components to plan for:
- Meta's messaging charges: businesses pay for template messages they initiate, priced by category (such as marketing, utility and authentication) and by country. Replies to customers within the 24-hour customer-service window are generally free. Meta revises its pricing periodically, so check current rates when you plan.
- Provider or hosting fees: for the platform connection, inbox and hosting.
- Chatbot build and AI usage: designing the conversation, training it on your content, integrations, and the AI model costs per conversation.
For most SMEs, the biggest variable is how many business-initiated messages you send. A support-focused bot that mainly replies to customers costs much less to run than a marketing-heavy one.
Rules You Must Follow
- Opt-in: customers must agree to receive messages from you before you send them business-initiated messages.
- Approved templates: messages you start (outside the 24-hour window) must use templates approved by Meta.
- Commerce and messaging policies: certain products and content are restricted.
- Data protection: under India's DPDP Act 2023, be clear about what you collect in chat and why, and protect it.
Best Use Cases for Indian SMEs
Retail and distribution
Stock and price questions, order status, delivery updates, catalogue sharing and payment reminders. A bot connected to your order system can answer "Where is my order?" with live information.
Clinics and diagnostic centres
Appointment booking and reminders, timings, test preparation instructions and report-ready alerts. Clinical questions should always go to staff, never be answered by the bot.
Education and coaching
Admission enquiries, course details, fees, batch timings and counselling bookings — often in Marathi or Hindi, at peak season volumes.
Real estate
Instant replies to portal leads, project details, price ranges, qualification questions and site-visit booking.
Manufacturers and B2B suppliers
Dealer order intake, order and dispatch status, document requests (invoices, test certificates) and service requests.
Hindi, Marathi and Hinglish Support
One of the biggest advantages of modern AI chatbots in India is language. Customers type in English, Hindi, Marathi, Romanised Hindi ("order kab aayega?") or a mix. AI models handle this far better than menu bots. The key is testing: collect real customer messages in each language, check the bot's replies with a fluent speaker, and tune the tone — formal for some businesses, friendly for others.
Human Handoff: The Feature That Makes or Breaks a Bot
Customers get frustrated when a bot traps them. Good handoff design includes:
- Clear triggers: complaints, pricing negotiations, urgent issues, or whenever the bot is unsure.
- An easy way for the customer to ask for a person at any time.
- Passing the full chat history to staff, so customers never repeat themselves.
- Telling the customer honestly when a person will reply, especially outside business hours.
How to Launch a WhatsApp Chatbot in Weeks
- List your top 20–30 customer questions from real chats — these usually cover most of your volume.
- Gather the answers and source documents: price list, policies, timings, FAQs.
- Set up the WhatsApp Business Platform with an approved provider and verify your business with Meta.
- Build the bot on your content, with handoff rules and your brand's tone.
- Test with staff in every language your customers use, then with a small group of customers.
- Launch, review conversations weekly, and add answers for questions the bot missed.
- Add integrations — order lookup, appointment booking, CRM — once the basics work well.
A focused FAQ and lead-capture bot is often live in two to three weeks; bots connected to order systems or calendars typically take a few weeks more.
Measuring Success
- Resolution rate: share of conversations the bot completes without a person.
- First response time: should drop from hours to seconds.
- Handoff quality: are escalations reaching the right person with context?
- Customer satisfaction: a simple thumbs-up or rating after resolution.
- Business outcomes: leads captured, appointments booked, staff hours saved.
WhatsApp Bot, Website Chatbot or Both?
The same AI and knowledge base can power your WhatsApp bot, your AI chatbot for your website and an internal helpdesk, keeping answers consistent everywhere. If your bot needs to answer from large or changing document sets, it can be upgraded with RAG so answers are always grounded in your latest content.
A Sample Conversation Flow
Here is how a well-designed bot for a coaching institute might handle a typical enquiry. A parent writes in Marathi asking about the Class 10 batch. The bot replies in Marathi with batch timings, fees and the start date, then asks whether they would like a counselling call. The parent asks about a scholarship; the bot shares the published scholarship criteria and offers a call with the counsellor. The parent agrees, picks a slot, and the counsellor receives the booking with the full conversation attached. No staff time was spent until a genuinely interested parent needed a person.
Common Mistakes When Launching a WhatsApp Bot
- Launching with long menus instead of letting customers type naturally.
- Hiding the option to talk to a person.
- Sending too many promotional messages, which annoys customers and raises costs.
- Letting answers go out of date when prices or policies change — assign someone to keep the knowledge base current.
- Not reviewing conversations after launch; the questions the bot misses are your best improvement list.
Finally, set expectations with your own team. Staff should know which questions the bot handles, how handoffs reach them and how quickly they should respond. When customers are passed to a person, a fast human reply is what turns a good bot experience into a good business experience.
Want to see a WhatsApp chatbot answering your customers' real questions? Book a free demo or learn more about our WhatsApp chatbot development in India.