How to build a WhatsApp chatbot without code: a coaching institute admission bot, step by step

What you are building

You can build a WhatsApp chatbot without code in Broadchat's visual flow builder: you place steps on a canvas, connect them with lines and fill in forms. No scripting, no developer. This guide builds one real bot from start to finish, an admission enquiry bot for a coaching institute, because a concrete example teaches the builder faster than a tour of every menu.

Our example is Shiksha Point Classes, a JEE and NEET coaching institute in Jaipur. Parents and students message the institute's number after seeing a newspaper ad or a hoarding with a QR code. Today two counsellors answer the same five questions all day: which course, which class, the fee, the batch timing and "can my child attend a demo". The bot will ask those questions, send the right fee brochure, save every answer on the contact, tag the demo requests and hand serious enquiries to a counsellor during office hours.

Step 1: plan the conversation on paper

Most bots fail because they try to do everything. Give this one a single goal: a qualified enquiry with a demo request, handed to a counsellor. Then list the questions and the exact options. Every question should change what the counsellor does next; if it does not, drop it.

WhatsApp itself sets the limits you design around: a message can carry up to three reply buttons with titles of up to 20 characters, and a list message can hold up to ten rows in total, with row titles up to 24 characters. That is why the course question is a list and the class question is three buttons.

Step 2: create the fields that will hold the answers

Answers are only useful if your counsellors can see them. Before you open the builder, go to CRM → Super Field and create three fields: course_interest, current_class and pincode (text fields are fine). Super Fields are custom contact fields: they show on the contact's page in CRM → Contact (and as columns in the contact list), can be used as conditions in segments, and can be inserted into messages as variables such as {{course_interest}}.

If you only need a value inside this one bot (for example a temporary choice), use Chatbots → Bot Fields instead. Bot Fields are variables the bot keeps; Super Fields belong to the contact.

Step 3: create the bot and set the trigger

1. Open Chatbots → Chatbots and click Create Botflow. The builder opens with a new bot called "Untitled"; click the name at the top left and rename it "Admission enquiry".
2. Click the first step on the canvas (the trigger). Under Bot Trigger, choose Match Keyword.
3. Under Conditions, build: Incoming Message Contains admission, then Add OR Condition: Incoming Message Contains course, and another OR: Incoming Message is hi.
4. Under Select Channel, add your WhatsApp number. If you leave it empty, the bot runs on all channels.

Matching ignores capital letters and extra spaces, so "Admission?" and "ADMISSION" both match. Put the same word in your QR code's prefilled text (for example Hi, I want admission details) so people who scan it start the bot without thinking.

Step 4: add the course question as a list

Click the round + button at the bottom right of the canvas to open the Quick Menu, or drag a line out of the trigger into empty space. Choose Ask Question, then:

1. Set the question type to Interactive. Body: Welcome to Shiksha Point Classes, Jaipur. Which course would you like details for?
2. Set Send Choices as to List. Button Text: View courses. Section Title: Courses 2026-27.
3. Add five items with + Add Item In Section, each with an Answer Label (24 characters at most) and an optional Answer Description, for example JEE 2027 Batch / Class 11 and 12, Mon-Sat.
4. Under Save Response to, pick the CRM Super Field course_interest.
5. Click Save Question.

Each list row now has its own output dot. This is how a no-code bot branches: you connect each answer to a different next step. The Ask Question step also has outputs for No Input (the person did not answer within the Response Waiting Time, 24 hours by default) and No Match (they typed something that is not on the list, after all attempts). Connect No Match to a short Send Message such as "Please tap View courses and pick one" and route it back with a Goto step.

Step 5: ask for the details you need, with validation

The four course rows will each get their own brochure in step 6, and the "Talk to a counsellor" row goes straight to the hand-off in step 7. The questions below are shared by all four courses, so build them once, as a chain:

• Student's name: type Text, Expected Answer Type (Validation) Text, saved to the Contact field Chat Name, so the counsellor sees the student's name instead of a number.
• Current class: type Interactive, Send Choices as Button, three buttons Class 10, Class 11, Class 12 / Dropper, saved to current_class.
• Pin code: type Text, validation RegExp with the expression ^[1-9][0-9]{5}$, Retry Message (If Validation Fail) Please send the 6-digit pin code, for example 302017, Validation Attempts Three Times, saved to pincode.

Every answer is also available as {{N.answer}}, where N is the number on the step's card, and the contact's name as {{name}}. The Mapping button ({ } icon) in the builder's top bar lists every variable you can copy.

Step 6: send the right brochure for each course

Fee brochures differ by course, so this is where the separate list outputs pay off. Connect each course row to its own Send Message step of type Document, upload the PDF (or paste a public URL), set the File Name to something readable like Shiksha-Point-JEE-2027-fees.pdf, and add a caption:

Then connect every brochure step to the first question of the chain from step 5 (the student's name). The result is four short branches that rejoin, which keeps the flow readable. Keep media within WhatsApp's limits (documents up to 100 MB, but a 1-2 MB PDF opens much faster on a parent's phone).

Step 7: tag the lead and hand it to a counsellor

Add a final Ask Question of type Interactive with three buttons: Yes, book a demo, Call me first and Not now. Then wire each answer:

1. Yes, book a demo and Call me first: add Add Automation App (the dark button at the bottom of the Quick Menu), choose the CRM app and Assign Tags to Contact (Creates Tag if Tag not exist) with the tag demo-requested or callback. Next, add Working Hours (Native Tools) with Time Zone Asia/Kolkata and Monday to Saturday, 09:00 to 19:00.
2. Inside working hours, add Assign to Agent, choose Send to Department and pick your Admissions department. Broadchat gives the chat to the active member of that department with the fewest open chats. Add a Send Message: Thank you {{name}}. A counsellor from our admissions desk will message you here in a few minutes.
3. On Out of Working Hours, send Our admissions desk is open Monday to Saturday, 9 am to 7 pm. A counsellor will message you first thing tomorrow. and still assign the chat, so it is waiting in the counsellor's queue in the morning.
4. After both branches, add Stay in Session. It keeps the contact inside this bot for the rest of the 24-hour session, so another bot does not jump in while the counsellor is chatting.
5. Not now: tag nurture, send No problem. Type ADMISSION any time to see the courses again. and end the flow.

Step 8: save, switch on and test from your phone

Saving a step's panel only updates the canvas; you will see "Step updated - remember to Save Workflow". Click Save Workflow at the top right (an amber dot means unsaved changes), then turn on the Active switch next to it or in the chatbot list.

There is no simulator in the builder, so test the real thing: from a phone that is not the business number, send admission. Walk every branch at least once, including a wrong pin code and a "Not now". Then open CRM → Contact, find your test number and check that the Super Fields and tags are filled, and open CRM → Live Chat to confirm the chat was assigned to the department. If a step fails, open the bot's Error Log from the chatbot list; it shows the time, contact, step and error.

After launch: what it costs and what to improve

Bot replies to a customer who messaged you are service messages inside the 24-hour customer service window: each business phone number gets 1,000 free service messages per month, and after that Meta charges per message (₹0.115 per service message to India, Meta rate card effective 1 Oct 2026, excl. GST). If the bot or a counsellor needs to follow up after 24 hours, you need an approved message template; see the lead follow-up templates. Each Add Automation App step (such as the CRM tag) also uses a task from your plan.

After two weeks, look at where people stop. If many abandon at the pin code, make it optional or move it after the brochure. If counsellors still ask the same question, add it to the bot. For the wider admissions playbook, including fee reminders and parent notices, read WhatsApp API for education, and for more flows like this one see the chatbot builder. Agencies that build bots like this for schools and coaching centres can sell them under their own brand through the white-label partner program.