WhatsApp campaign analytics and delivery tracking: the metrics that actually matter

Most WhatsApp campaign dashboards answer a question nobody asked. They show you that 41,882 messages were sent and 39,104 were delivered, and then they stop — leaving the person reading it with a large number, a vague good feeling, and no idea what to do differently on Monday.

Useful WhatsApp analytics is not about collecting more events. It is about reading the gaps between the events you already have.

The four states, and what each gap tells you

Every message moves through a small set of states, and each transition that fails to happen points at a different problem.

Sent → Delivered

This gap is a list-quality metric. Sent means the API accepted the message. Delivered means it reached a device. Messages that never arrive are usually numbers that are inactive, are not on WhatsApp, or have blocked you.

Read it per segment, never in aggregate. A 94% delivery rate across a campaign can easily be 99% on your recent buyers and 61% on a segment somebody imported from a spreadsheet in 2024. The aggregate hides the exact thing you needed to find, and that stale segment is also the part generating the blocks that move your quality rating.

Delivered → Read

This gap is content and timing. It arrived; they did not open it. Test the opener, the sender identity and the send hour here — not the offer, which they have not seen yet.

One caveat worth knowing before you build a target around this: read receipts can be disabled by the recipient, so read rate is a floor rather than a precise count. It is comparable across your own campaigns. It is not comparable to someone else’s published number.

Read → Replied or Clicked

This gap is the offer. They read it and did nothing. No amount of subject-line work fixes a weak offer, and this is the only metric in the list that is genuinely about your commercial proposition rather than your plumbing.

The fifth state nobody puts on the dashboard

Blocks and reports. This is the metric that should sit at the top of the report, because it is the only one with consequences beyond the campaign. Blocks and reports drive your number’s quality rating over a rolling 7-day window, and a Low rating stops your account advancing to a higher daily messaging tier.

A campaign that produces 3% more revenue and doubles your block rate is a loss, and a dashboard that reports only the first half of that sentence will let you run it again.

The report that actually changes behaviour

In practice a WhatsApp campaign report needs five rows, split by segment:

1. Delivery rate by segment — finds the dead lists.
2. Read rate by segment and send hour — finds the wrong time and the wrong opener.
3. Reply or click rate by template — finds the weak offer.
4. Block and report rate by segment — finds the audience that is quietly costing you your messaging tier.
5. Cost per outcome — conversations are billed, so the honest denominator is money, not messages.

Everything else — total sent, total delivered, cumulative recipients — is a vanity row. It is fine to have. It should not be the first thing on the page.

Attribution without deceiving yourself

WhatsApp attribution is genuinely harder than email, because a meaningful share of conversions happen inside the conversation itself. Someone reads a campaign, replies, talks to an agent, and buys — with no link click anywhere in the chain. Click-only attribution will report that campaign as a failure.

Two practical corrections:

• Count replies as a conversion event in their own right, not merely as engagement. On WhatsApp a reply is much closer to the purchase than a click is on email.
• Tag the conversation, not just the link. If an agent closes a sale in a thread that began with campaign X, the revenue belongs to campaign X even though no URL was ever opened.

And one thing to avoid: quoting a generic industry open-rate figure as though it were your result. Those numbers circulate without traceable sources, and the moment a colleague asks where it came from, everything else in your report gets re-examined too. Your own measured read rate is a smaller number and a much stronger position.

What we built into the product

AI WhatsApp by Vfastrr reports per-template delivery, read, reply and conversion events, split by segment rather than rolled into one campaign total — and it connects that to number-quality intelligence, so the block rate a campaign generated is visible next to the revenue it produced instead of surfacing three days later as a colour change in Business Manager.

The point of campaign analytics is not to prove the campaign worked. It is to know which part to change.

Related

• WhatsApp number quality rating dropped: why it happens and how to fix it
• WhatsApp daily messaging limit exceeded