Skip to main content
Waaru
WhatsApp for D2C ecommerce

Answer product questions on WhatsApp before a customer buys

Tanay PatelAI and full-stack engineer
The short answer

Answer the product question first, using current information for the exact item or variant. Ask a follow-up only when it changes the recommendation. Be clear when fit, stock or delivery cannot be confirmed and give the customer a useful next step.

A customer asking “Will this fit?” may be one detail away from a decision. Sending the entire catalogue shifts the work back to them. A better reply narrows the uncertainty without pretending to know more than the product information supports.

How do you find out what the customer means?

Start with the product and the decision. A photo, a short name or “the blue one” may not identify a unique item. Ask for a product link or variant when needed. Do not assume the most recently promoted item is the one they mean.

Fictional example: A customer asks whether a bag fits a laptop. A useful reply is: “Which laptop dimensions should we check, and which bag are you considering?”

The answer can then compare the published compartment dimensions. It should not guarantee fit from screen size alone when the casing or shape may differ. Explain the missing information in ordinary language so the customer understands why you asked.

What do helpful replies look like?

These are illustrative replies to adapt after checking your own product information.

Size: “The size guide lists the garment measurements on this product page. Which measurement would you like help comparing?”

Compatibility: “The specification lists these supported models. Your model is not listed, so I need the product team to confirm before recommending it.”

Material: “This variant is listed as cotton with a polyester lining. The care instructions are on the linked page.”

Each reply answers what can be answered and names the remaining question. Avoid phrases such as “perfect for everyone” or “definitely suitable” when you have no basis for them. A customer can make a better decision from an honest limitation than from a confident guess.

Which answers need a fresh check?

Stock, current prices and delivery information can change. A saved answer about a product’s material may remain useful, while a saved claim that a colour is available can quickly become wrong.

Use this product-answer checklist:

  • Exact product and selected variant.
  • Source of the specification or policy.
  • Whether the information changes frequently.
  • Information needed to answer this customer’s question.
  • Any limitation the customer should know before buying.
  • The owner who can resolve an exception.

If a stock check fails, say it needs checking. Do not substitute the last successful result without explaining its age and limits. A catalogue description is not necessarily live inventory.

What should a person decide?

Send uncertain compatibility, special-order requirements and exceptions to a qualified product or support teammate. Give them the item reference and the customer’s actual question. Do not summarise “not sure which one fits” as “ready to purchase”.

The human handoff guide explains how to avoid making the customer repeat the conversation. Keep the same ownership discipline for returns-policy exceptions and delivery commitments.

If an AI answer uses a knowledge base, the source still needs review. The knowledge-base guide covers preparing material for that purpose. A fluent answer does not establish that a specification is correct or current.

How do you help without pushing an unrelated offer?

Once the question is answered, offer the relevant product page or ask whether another detail would help. Do not turn every answer into a demand to buy immediately. Someone may need to compare measurements or check with another person first.

Keep later outreach within WhatsApp’s permission and messaging requirements. A product question does not authorise an unrelated promotional sequence. The opt-in guide explains the distinction.

If the customer has already bought and asks where the item is, switch to the order-status guide. Recommending another product does not answer a delivery question.

What should you prepare before adding automation?

Collect actual question categories from appropriately authorised support material, then write approved answer sources. Use fictional examples during testing. Include an unknown item, a missing specification and a failed stock lookup. Review whether the reply admits uncertainty and offers a workable handoff.

The D2C page connects product questions to customer support and order workflows. Use the glossary to explain terms such as knowledge base and integration to your team. Start with the questions you can answer reliably before expanding the scope.

FAQ

Frequently asked questions.

Can a saved reply answer every product question?

No. Saved information can help with stable specifications, but the exact variant and changing details still need checking.

Should an AI agent recommend a product when information is missing?

It should identify the missing detail or route the question to a person. Do not treat an invented compatibility or suitability answer as a recommendation.

Research notes

Evidence checked for this article.

  • WhatsApp permission, templates and customer-service messaging requirements.

    WhatsApp · Accessed 2026-09-16

About the author

Tanay Patel

AI and full-stack engineer

Tanay builds production AI systems, agent workflows, and the full-stack software around them. His work focuses on reliable state, recovery, and human handoff.

Read Tanay's engineering notes

Put this into practice on your own number.

The platform is free. AI features are paid separately.