Click-to-WhatsApp ads: what happens after the click

A Click-to-WhatsApp advertisement runs on Facebook or Instagram and, instead of sending the click to a website, opens a WhatsApp conversation with the business. The person arrives in a chat with a message already typed into the input field, ready to send.

The reason this format behaves differently from ordinary advertising is that it collapses several steps. There is no landing page to load, no form to complete, no email address to type. The distance between seeing an advertisement and being in a conversation with a person is a single tap, on an application the customer already has open and already trusts.

It also inverts who starts the conversation, which has consequences beyond convenience. Because the customer sends the first message, the customer service window opens, and the business can reply freely in its own words rather than through an approved template. Meta treats arrival through this route as a qualifying entry point, so the resulting conversation does not carry the cost profile of an outbound template send. The advertising spend is real and sits on its own line, but the conversation itself begins on favourable terms.

The format's weakness is the same as its strength. It generates conversations, and conversations require somebody to be there.

The text waiting in the input field is the single most consequential setting in the whole setup, and it is routinely left as a greeting. "Hi" tells the business nothing. A conversation that opens with "Hi" requires the business to ask what the person wants, wait for an answer, and only then begin being useful — two messages spent establishing what the advertisement already implied.

A better pre-filled message states the specific thing the person is enquiring about, phrased as something they would plausibly have typed. If the advertisement is about a particular service, the pre-filled text should name it. This gives the business the context to make its first reply substantive, and it gives the customer the sense of having been understood rather than processed.

It should also be short and editable. The customer can change the text before sending, and a long paragraph written in the business's voice reads as a script being put into the customer's mouth, which some people will delete rather than send. One line, in plain language, phrased as a question the person actually has, gets sent as written far more often.

Where several advertisements run at once, varying the pre-filled text by advertisement is how a business can tell which one produced an enquiry without instrumenting anything further.

An advertisement that works produces conversations arriving at unpredictable times, and where they land determines whether they are answered. This is a decision to make before the campaign runs, not after the enquiries start accumulating.

The questions are mundane and important. Which number receives these conversations. Who is watching that number, during which hours. What happens to a conversation that arrives outside those hours. Who takes over when the first person is unavailable, and how they know what was already said. A business where one person answers everything on their personal handset has an answer to all of these, and it stops working the moment that person is asleep or busy.

The hours question deserves particular attention with paid advertising, because advertisements often continue delivering when nobody is available to respond. A business either restricts delivery to hours it can cover, or accepts overnight enquiries and sets an expectation for when they will be answered. Paying to generate a conversation that nobody attends to is the most straightforward way to waste the budget.

It is also worth deciding what to record. An enquiry that arrives, gets answered and disappears into a chat history is a lead the business cannot follow up. Knowing which advertisement produced it, what was discussed and what happens next is the difference between a channel and a series of unconnected conversations.

The advertisement bought attention; the first reply decides what happens with it. Someone who has just tapped an advertisement is, at that moment, more interested than they will be at any later point. They are also on WhatsApp, where the ambient expectation is a prompt reply, not the slower rhythm of email.

This is not just a matter of expectation — the mechanics reinforce it. The customer's message opened a 24-hour window during which the business can reply in its own words. Let the window close and the same conversation requires an approved template, which is a stiffer instrument for what was a live enquiry. Speed here is not politeness; it is the difference between a conversation and a re-approach.

The reply should do something rather than acknowledge. "Thanks for your message, how can I help?" wastes the context the pre-filled message provided. If the person enquired about a specific service, the useful reply answers the obvious next question — availability, price range, what happens next — and asks one thing the business genuinely needs to know.

If an immediate human reply is not possible, an honest holding message that states when someone will respond is better than silence. What does not work is a holding message that pretends to be a person, because the follow-up question exposes it and the customer has then been misled as well as delayed.

Click-to-WhatsApp campaigns have two costs that behave differently, and conflating them makes the economics impossible to read. The advertising spend buys the click and is priced by the advertising auction. The messaging cost is governed by WhatsApp Business Platform billing and depends on which messages the business subsequently sends.

Because the customer initiated, the conversation starts as a free-form exchange inside an open window rather than as a paid template send. That is favourable, and it is also why the messaging cost of this channel is easy to underestimate in the other direction: any follow-up sent after the window closes is a template, in whatever category matches its content, and a sequence of follow-ups is a sequence of charges under per-message billing.

There is a third cost, unbilled and usually the largest: the staff time to answer. A campaign that generates far more conversations than the business can attend to has not succeeded. Judging performance on cost per conversation alone hides this, because the metric improves precisely as the conversations become less well handled. Cost per conversation that resulted in something — an order, a booking, a qualified follow-up — is the number that reflects reality.

A short readiness check prevents most of the waste. Is the pre-filled message specific to the advertisement, short, and phrased as something the customer would type. Is the receiving number one somebody is actually watching. Are the advertisement's delivery hours aligned with the hours somebody can reply, or is there an honest holding message for the gap. Does the business know what it will say first, given the context the pre-filled message supplies. Is there somewhere an enquiry gets recorded so it can be followed up after the chat scrolls away.

It is also worth deciding in advance which follow-ups will be needed once the window closes, and getting those templates approved before the campaign starts. Discovering that a follow-up requires an approved template, and that approval takes time, while warm enquiries go cold is an avoidable and common sequence.

Starting small is the right default. Run the campaign at a spend the business can comfortably service, watch how many conversations arrive and how many get answered well, and increase only once the answering side keeps pace. The constraint on this channel is almost never the advertisement's ability to generate conversations. It is the business's ability to have them.