English

Click-to-WhatsApp Ad Campaigns: From Setup to First Conversion

A performance marketer ran the same creative two ways: once as a traffic campaign driving to a landing page, and once as a Click-to-WhatsApp campaign. The traffic campaign brought in leads at $42 per qualified conversation. The Click-to-WhatsApp campaign dropped that to $7. Same creative. Same audience. Six times cheaper per qualified conversation. Meta's own advertising data shows that Click-to-WhatsApp campaigns consistently outperform link-click campaigns on cost-per-conversation for warm B2B audiences.

The difference wasn't the ad. It was the setup. Click-to-WhatsApp campaigns require specific decisions at the objective, message template, and automation handoff levels that traffic campaigns don't. Get those wrong and you're paying for opened chats that go nowhere.

This guide walks through the five-step setup in Meta Ads Manager, from campaign creation to your first qualified conversation in the inbox.

How Click-to-WhatsApp Differs From Other Meta Objectives

When someone clicks a Click-to-WhatsApp ad, Meta opens WhatsApp directly on their device with a pre-filled message ready to send. The broader shift away from form-based lead gen is explored in why the contact form is dying as a B2B lead capture method. The moment they tap Send, the conversation begins: no landing page, no form, no email address required.

This is fundamentally different from Lead Gen ads (which collect data inside Meta) and Traffic campaigns (which send people to a URL). With Click-to-WhatsApp:

  • The conversion event is the conversation start, not a form submission
  • Meta's delivery algorithm optimizes for people likely to open WhatsApp and send the pre-filled message
  • The "funnel" happens inside WhatsApp, driven by your automation flow
  • Attribution is conversation-based, not click-based

The implication: if you set up the campaign as if it's a traffic or awareness objective, Meta won't optimize delivery correctly and your cost-per-conversation will be 2-3x higher than it should be.

Step 1: Campaign Setup in Meta Ads Manager

  1. In Meta Ads Manager, click Create → New Campaign
  2. Select Messages as your campaign objective (not Traffic, not Engagement)
  3. Under "Messaging Apps," select WhatsApp
  4. Name your campaign with a convention that includes the product, audience, and date (e.g., "CRM-RetargetWebVisitors-Apr2026")

Phone number requirements. Your WhatsApp Business number must be connected to a verified Meta Business Manager. If it isn't, the Messages objective won't show WhatsApp as an available destination. Verify under Business Settings → WhatsApp Accounts.

Business Manager permissions needed:

  • WhatsApp Business Account access (Admin or Editor)
  • Ad Account access (Advertiser or above)
  • Page access for the connected Facebook Page

If you're running ads for a client, they need to grant you these via Business Manager partner access, not just ad account access.

Step 2: Crafting the Opening Message Template

When someone clicks your ad, Meta pre-fills a message in their WhatsApp input field. This text is your opening message template. The user just taps Send.

The quality of this text determines whether people actually send it. A message that feels natural gets sent. One that feels like a form submission gets abandoned.

Opening message text Send rate signal Why it works (or doesn't)
"Hi, I'm interested in learning more" Low Generic, no reason to send it
"Hi! I saw your ad and want to know more about pricing" Medium More specific but still passive
"Hey, can you show me how [Product] handles [specific use case]?" High Expresses a real question the user already has
"I want to see the demo for sales teams" High Specific intent, feels like something a real person would say
"Send me the guide on [topic from ad]" High Exchange-based: they're expecting something in return

The pattern that consistently works: tie the message text to a specific thing promised in the ad creative. If your ad says "Get the free sales benchmarks report," the pre-filled message should say "Send me the sales benchmarks report." The user's expectation matches what they're about to send.

Template approval note. Some message template types require Meta review before they can be used. For Click-to-WhatsApp ads using pre-filled messages, the text you write at the ad set level doesn't require separate approval. It's part of the ad. But if you're sending outbound WhatsApp messages using the API (not the user-initiated flow), those require template approval. Know which you're doing.

Step 3: Audience Targeting for Chat

Not all audiences are equally likely to engage via chat. The format rewards warm audiences and high-intent segments.

Audiences that work well:

  • Website visitors from product or pricing pages (last 30-60 days)
  • Lookalike audiences modeled on your existing WhatsApp converters
  • Interest segments where the audience has demonstrated high search/click intent on your category
  • Email list custom audiences (warm contacts who haven't converted)
  • Video viewers who watched 75%+ of a product demo video

Audiences that waste budget:

  • Broad interest targeting with no intent signal
  • Top-of-funnel awareness audiences (they don't know you yet and won't start a chat cold)
  • Audiences that performed well for lead form campaigns (lead form behavior doesn't predict chat behavior)

Bid strategy. Start with Advantage+ placements and Highest Volume bidding for the first few days. Once you have 50+ conversations, you can switch to Cost Per Result Goal with a target cost per conversation to cap spend. Statista reports that WhatsApp has over 2 billion monthly active users globally, making it one of the highest-reach channels for mobile-first B2B audiences in international markets.

Step 4: The Automation Handoff

This step is where most Click-to-WhatsApp campaigns silently fail. Once the automation flow is live, you'll also want to think through your conversational qualification sequence — because how you structure questions after the first message determines whether the lead continues or drops off. The ad starts a conversation, but if no automation flow is triggered, the only response the lead gets is silence. Or worse, a rep manually responding at midnight.

Connecting to ManyChat. In ManyChat, create a new WhatsApp flow with a keyword trigger matching the pre-filled message text from your ad. If your pre-filled message is "Send me the sales benchmarks report," set a keyword trigger for "sales benchmarks." Add a fallback trigger for common variations.

Alternatively, use ManyChat's Meta ad integration: under Growth Tools → Click-to-WhatsApp, connect the specific ad to a flow. This is more reliable than keyword matching because it fires based on the ad source rather than message text.

Connecting to Respond.io. In Respond.io, use the Inbox Automation → Assignment Rules to route conversations that arrive from a specific ad source to a specific flow or team. You can identify the ad source via the conversation metadata that WhatsApp passes through (campaign ID appears in contact attributes).

The critical check. Test the full path: click the ad on your own phone, send the pre-filled message, and confirm the automation flow fires within 5 seconds. If it doesn't, the ad-to-flow handoff is broken.

Step 5: Tracking and Attribution

UTM parameters in WhatsApp context. Standard UTM parameters don't apply to WhatsApp conversations the way they do to website visits. The conversation doesn't carry a URL. Instead, attribution works via:

  1. Meta's in-conversation attribution (shows as "Conversations Started" in Ads Manager, attributed to your ad)
  2. CRM-side attribution using the campaign ID captured from the WhatsApp conversation metadata
  3. A manual "source" question in your qualification flow ("How did you hear about us?"), simple but surprisingly reliable

What "cost per conversation" really means. Meta reports "conversations started" which means the user sent the initial message. This includes people who open the chat and send the message but then immediately go silent. Your real metric is cost per qualified conversation, where "qualified" means the person completed your flow and hit your minimum qualification threshold. Track this in your CRM, not in Meta. A Forrester research report on conversational marketing found that B2B teams separating conversation-starts from qualified conversations see 2-3x more accurate pipeline forecasts.

In-app vs CRM-side attribution. Meta Ads Manager will show you conversations started, conversation rate, and cost per conversation. Your CRM will show you which conversations became leads, meetings, and deals. For a deeper look at why standard attribution breaks down for chat channels, see why attribution is broken for conversational revenue. You need both views. Don't try to optimize the Meta campaign on CRM data alone (the feedback loop is too slow) or on Meta data alone (it overstates quality).

Scaling What Works

After 3 days of running, look for these signals before scaling. Your chat funnel metrics dashboard should have the completion rate and cost-per-qualified-conversation figures you need to make the call.

Scale signals:

  • Completion rate (conversation fully qualifying) above 40%
  • Cost per qualified conversation below your target (compare to lead form CPL)
  • A meaningful percentage of completions converting to pipeline in the CRM

Kill signals:

  • Conversation start rate below 30% (the pre-filled message isn't being sent: message text or audience mismatch)
  • Drop-off after first bot message above 70% (the flow isn't holding attention)
  • No CRM contacts being created (integration broken)

When you find a winner, increase budget by 20% every 48 hours to avoid resetting Meta's learning phase. Don't double spend overnight.

Common Pitfalls

Using the wrong campaign objective. Running Click-to-WhatsApp ads under Traffic or Engagement objectives means Meta isn't optimizing for conversations. You'll get clicks that don't result in opened chats.

Message template creating a silent ad failure. If your pre-filled message triggers Meta's spam detection (unusual characters, all caps, suspicious phrasing), Meta may reject the ad silently, so it runs but no conversations start. Check the ad's delivery status under Ads → Columns → Performance and Clicks → Quality Ranking.

No automation flow connected. The most expensive mistake. Reps manually responding to 50 conversations a day from a scaled campaign is unsustainable and creates inconsistent follow-up. Wire the flow before you run the ad.

Double-counting in Meta vs CRM. Meta counts a "conversation" the moment the pre-filled message is sent. Your CRM counts a lead when qualification is complete. The gap between these two numbers is your funnel drop-off rate. Don't confuse the two in your reporting.

What to Do Next

Run this Click-to-WhatsApp campaign in parallel with a standard Lead Gen campaign targeting the same audience with the same creative for 2 weeks. Once you have enough conversation volume, chat funnel A/B testing gives you a structured framework for optimizing the pre-filled message and automation flow based on real data. Compare cost-per-qualified-lead across both. For most B2B audiences, Click-to-WhatsApp will win on warm audiences and Lead Gen will win on colder ones. That data tells you exactly where to allocate budget going forward.

Learn More