How to build a WhatsApp chatbot for your online store

Start from what customers actually send. Shoppers ask about delivery time to their pin code, cash on delivery, size and fit, return policy, and where their order is.

Then look at what goes wrong today:

• 68% of carts abandoned with no follow-up.
• Order tracking calls overload your support. "Where is my order?" is your most common support query. WhatsApp order updates answer it before customers ask.
• COD orders come back as returns. The order ships, the customer changes their mind, and you pay both legs of the shipping plus the packaging.

Each of these is a conversation with a predictable shape — a question, a lookup, a confirmation. That shape is what a chatbot handles well.

1. Customer types "where is my order" or taps [Track order].
2. The bot recognises the number and shows recent orders, or asks for the order number.
3. It shows the status and tracking link, and the expected delivery date.
4. If the order is late, it says so, gives the courier's latest update, and offers to escalate.

Notice what the business side does: it answers with the specific detail (time, reference, next step), it asks for one thing at a time, and it gives the customer a one-word way to reply. Those three habits matter more than how clever the bot is.

The bot asks which order and item, the reason (size, damaged, not as described), photos where needed, and whether the customer wants a refund or exchange. It checks eligibility against your policy — within the return window, item category — and schedules a pickup or explains why it cannot. Exceptions go to staff.

Reputoo supports both, and most online stores end up using both:

• Keyword flows — a fixed menu or trigger word ("MENU", "BOOK", "STATUS") leads down a set path. Predictable, easy to audit, right for anything with a fixed answer.
• AI chatbot — reads the customer's own words and answers from your knowledge base (website, price list, policies). Right for the long tail of questions nobody wrote a menu item for.

A sensible start for an online store: keyword flows for the top two or three tasks, AI for everything else, and a person behind both.

"Will this fit a 32-inch waist?", "is it cotton?", "can you deliver to 411045 by Friday?" — the bot answers from product details and the delivery-time rules, and sends the product link with an [Add to cart] route. Questions the product details do not answer go to staff quickly, because a shopper waiting for an answer is a shopper about to leave.

For stores with many products, a short product finder — who is it for, occasion, budget — narrows the catalogue to a few suggestions with photos and links. It works best for gifts, where shoppers do not know exactly what they want.

• Complaints, refunds and anything where the customer is upset.
• Anything involving money the bot cannot see — a disputed bill, a custom quote.
• Health, safety or legal questions.
• Any customer who asks for a human. Do not make them ask twice.

For online stores specifically: damaged or wrong deliveries, refunds and payment failures need a support person with access to the order.

Handover on Reputoo moves the chat into the shared inbox with the conversation so far, so staff never ask the customer to repeat themselves.

1. Connect your WhatsApp number to Reputoo (official API; takes minutes).
2. Add your website and documents to the knowledge base so the AI answers in your online store's own terms.
3. Build keyword flows for the top tasks above. What to connect: the store platform — Shopify, WooCommerce or a custom stack — is the source of orders, carts and customer consent.
4. Choose who gets handed-over chats, and during which hours.
5. Test from a personal phone, then share the number or a QR code with customers.

Before dispatch, the bot can change the address or cancel an order if your platform allows it, and confirm. After dispatch it explains the options. Changes it cannot make go to staff.

"Money deducted but no order" is a stressful message. The bot collects the time, amount and payment reference, checks whether an order exists, and hands to staff with everything attached, telling the customer when to expect an answer. It never asks for card numbers, PINs or OTPs.

Customers ask whether a code works or why it failed. The bot checks the offer rules and explains — minimum order, excluded items, expiry. It never creates codes; that stays with your team.

Shoppers write in many languages and mixes. Keyword flows cover common words — order, track, return, refund — and the AI chatbot handles the rest in the customer's language.

• Track a shipped order, a delayed order and a delivered one.
• Start a return inside and outside the window.
• Ask a size question and a delivery-time question.
• Report a payment deduction without an order and check the handover.

"Do you deliver to 781001?" is asked before many orders. The bot checks the pin code against your courier's serviceable areas and delivery times, and says whether COD is available there. A clear answer before checkout prevents orders that can never be delivered.

Shoppers buying gifts want gift wrap, a message card, delivery to someone else and no price on the invoice. The bot asks for these, confirms them on the order, and sends the tracking link to the buyer rather than the recipient unless asked.

For subscription products — coffee, pet food, skincare — the bot lets customers skip a delivery, change the date or product, or pause, within your rules, and confirms each change. Cancellations go to staff if you want a chance to save the subscription.

Handovers arrive with the order, the customer's words, photos, and the reason — damaged item, refund dispute, payment problem. The bot tells the customer when to expect a reply. Support staff start with everything they need.

Watch the share of "where is my order?" questions answered without staff, returns started in chat, pre-purchase questions answered before checkout, and support volume per order. Read handovers weekly for gaps in product details or policies.

During big sales, support messages multiply. The bot absorbs order tracking, delivery times and offer questions, so staff can focus on payment problems and exceptions. Update offer rules before the sale starts.

Size is the most common reason for fashion returns. The bot can offer an exchange first — "would you like the next size up instead?" — check stock, and arrange a pickup and a replacement together. Customers keep the product they wanted, and the store keeps the sale.

Customers reply to delivery messages with instructions — "leave it with the guard", "call before coming". The bot records these and passes them to the courier where your integration allows, or to staff to relay.

For electronics and appliances, the bot collects the order, the problem, photos or a video, and checks the warranty period, then routes to the right service process or brand service centre as your policy says.

When a product or size is out of stock, the bot offers to tell the shopper when it returns. With a tap they agree, and the alert goes out the moment stock arrives — one of the best-converting messages a store sends.

Orders abandoned at the payment step can get a fresh payment link on request, or once as a reminder to shoppers who opted in. The bot confirms when payment arrives and the order moves to dispatch.

A day after delivery, a short message — "Did everything arrive as expected?" with [All good] [There's a problem] — catches damaged or wrong items early. "There's a problem" goes straight to the returns flow with photos. "All good" is the moment to ask whether the customer would like to hear about new arrivals.

Customers ask how to wash, store or use what they bought. The bot sends the care instructions for that product from the catalogue — fewer damaged items, fewer returns, and a customer who feels looked after.