How long does a customer wait before giving up and buying from someone else? For most Indian small businesses the honest answer is: until the shop reopens. A boutique's WhatsApp pings at 9:40pm, a regular asking if a specific print is still in stock before a wedding this weekend, and nobody sees it until the shutter goes up the next morning. By then she has already messaged two other stores and bought from whichever replied first.
The problem is structural, not a matter of anyone slacking off. The store has exactly one person who can answer WhatsApp, and that person is also cutting fabric, running the till, and closing up for the night. Because the reply queue depends entirely on when a human happens to be free, the fastest-moving customers land on whichever business answers first, not necessarily the one with the better product or the better price.
The stakes are bigger than any single message. 78% of Indian SMBs already run some part of their business over WhatsApp, and customers now rank response speed as the single most important factor in a support interaction, ahead of how well the issue actually gets resolved. Businesses that reply within the hour keep about 71% of the customers who write in; those that take a day keep roughly 48%. That's not a rounding difference. It's close to a quarter of every conversation, gone, because the answer arrived too late to matter.
What speed is worth
A chatbot doesn't fix this by itself. Bolt one on with no boundaries and it either mishandles a complaint it was never built for, or a customer figures out in ten seconds they're talking to a script and gets more annoyed than if they'd just waited until morning. The versions that actually work are narrow on purpose: they catch stock checks, price questions, and order status the moment they land, at any hour, and route anything with real judgment in it, a complaint, a custom order, a price negotiation, straight to a person, with the full conversation history attached so nobody has to repeat themselves.
For the boutique in the opening, that would have meant one thing: a reply confirming the print is in stock, or naming the closest alternative, the moment the message came in at 9:40pm on a Tuesday, without anyone touching a phone. The business didn't need a smarter salesperson. It needed the answer to arrive before she opened a competitor's chat instead.
Start smaller than a full rebuild of how the shop handles messages. Pick the one question the team answers most often every single day, and build the response for just that question first, before touching anything else.
Where to start