How one WhatsApp keyword routes three different customer needs

Customers who message a brand for the first time don't all want the same thing. Some heard about a new launch. Some are browsing and would like help without committing to anything. Others already have a problem, such as a late order or a damaged item, and simply want it fixed. Most brands answer all of them with one generic welcome message, which usually means none of them is served especially well.

Ample Imagine built its inbound WhatsApp flow to separate those needs from the very first message. When a customer sends "welcome" or a similar keyword, the flow captures their name if it isn't already on file, sends a branded welcome card, and offers three paths: New Launches, Browse Products and Support Issues.

New Launches leads to Apple pre-order SKUs, with product-specific Buy Now templates for whatever is currently open for pre-booking. Browse Products hands the customer to an AI agent built for open-ended discovery, for people who know they want something from Apple but haven't decided what. Support Issues opens a nested menu with Track Order, Cancel, Return or Refund, Damage capture with a photo upload, and a direct Talk to an Agent option for anything that doesn't fit elsewhere.

The support branch does more work than it might appear. A customer who arrives with a problem and receives a generic welcome instead of a way to solve it tends to get frustrated quickly. Ample Imagine's flow gets that customer to the right sub-menu in the same first exchange in which a browsing customer gets a product recommendation. The entry point is the same, but the outcomes are tailored to each intent.

The flow also acts as the default catch-all for any inbound keyword that doesn't match a more specific trigger elsewhere in Ample Imagine's setup. Depending on the path a customer takes, it serves as an awareness entry point for launch discovery, a conversion assist for high-intent pre-book buyers, or an adoption touchpoint for order support. The customer's first reply decides which role it plays, so the brand never has to guess.

There's a fallback for support as well. If a ticket goes quiet, the flow closes it automatically after a silent window instead of leaving it open indefinitely, which keeps the support queue free of dead conversations nobody is working on.

In 30 days, 4.6K inbound welcome triggers were routed across the three paths, reaching 2.05K unique customers, with 98% delivery on 9.6K downstream messages. The delivery rate matters as much as the routing. A flow that sorts customers perfectly but fails to deliver one message in ten isn't solving the problem it was built for.

If your brand uses one generic welcome message for everyone who reaches out for the first time, this split is worth building. Identify the two or three real reasons people message you first, and let their first response route them.

See how this would work on your store

If you want to build a flow like this for your own store, book a meeting with the Manifest AI team and we will walk you through it in a live demo.