Skip to content
For media buyers

Know which source made money, not which one made volume.

A click carries through to the lead and the sale it produced, so every traffic source has a real margin — and Meta gets told about the conversion that actually happened.

Click-to-sale attribution
Blended EPC
Server-side Meta CAPI
Click tracking overview with clicks, conversions, EPC and revenue per traffic source
The problem

Sound familiar?

You can see the spend clearly. It is the return that goes missing.

Volume is not the same as money

The source with the most leads is often the one with the worst margin. Two dashboards later you still cannot say which campaign paid for itself yesterday.

You cannot see past the click

The ad platform tells you the click happened. Whether the lead sold, what it sold for, and what it cost you lives somewhere else entirely — and never in the same row.

The ad platform optimises against the wrong thing

A pixel that fires on the thank-you page counts form fills. It does not know the lead never sold, so it goes and finds you more people who fill in forms.

How Datahubb helps

Follow the money, not the traffic

The click, the lead, and the sale in one row

A click carries an ID through to the lead it generates, and the lead carries through to the sale. Revenue and payout attach to the click that actually produced them, so a source has a real margin rather than an impression count.

Both conversion points, counted once

A click can convert twice: it can produce a lead that sells, and it can trigger an advertiser CPC postback. Datahubb reports both — advertiser conversions and lead conversions, side by side — without adding them into the same total twice.

Blended EPC on the clicks you bought

Headline EPC divides conversion value by inbound clicks — the clicks you paid for, not the offer-side clicks downstream. That is the number that tells you whether the buy worked.

Meta CAPI fired server-side

When a click carries an fbclid, the conversion fires to Meta from the server, on the real outcome, with the fbc reconstructed and PII hashed to Meta’s rules. The platform then optimises against conversions that actually happened.

Rejection insights

Why a source’s leads do not sell

A source can send plenty of leads and still make you nothing, because the buyers keep rejecting them. Rejections are grouped by reason with the estimated revenue at stake against each — an expected value, not a guaranteed loss — and you can filter down to a traffic source, an affiliate ID, or a sub-source, so you find out whether it is a duplicate problem, a filter, or a buyer who was capped.

  • Rejection reasons ranked by the revenue estimated at stake
  • Filter down to a single source, affiliate, or sub-source
  • Cut the source, or fix the targeting — with a reason either way
Rejection reasons dashboard showing why leads were rejected and the revenue lost to each reason
What it looks like day to day

Judge a source on what it returned

Track a click from the ad through the form to the sale, on one ID
Read EPC, conversions, and revenue per source without exporting anything
See which source’s leads get rejected, and for which reason
Send server-side conversions to Meta on the outcome, not the form fill
FAQ

Frequently asked questions

Yes, and this is the point. A click can generate a lead that gets sold, and it can also trigger a CPC postback from an advertiser. Both are reported — the click statistics split advertiser conversions from lead conversions — and neither is double-counted, because lead revenue is only ever counted from the sale record and advertiser revenue only from the postback. See click tracking.

The headline blended EPC divides conversion value by inbound clicks — the traffic you actually bought. Offer EPC is a separate number that divides offer revenue by offer-side clicks, which is the publisher’s framing rather than yours. They are kept apart on purpose.

Because the conversion is fired from the server on the real outcome, not from a pixel on the thank-you page. If the click carried an fbclid, Datahubb sends the event to the Conversions API with the click context rebuilt and the personal data hashed to Meta’s specification. The platform then optimises for the leads that sold, not the forms that were filled. See conversions.

Some of them should not. Clicks are recorded with a status: valid, bot, or geo-blocked. Bot traffic is recorded but never redirected, so it does not inflate what the advertiser sees, and duplicate clicks from the same fingerprint are collapsed rather than counted. What is left is the traffic you can judge.

Find out which traffic actually paid.

Start your 14-day trial. Full platform access. Card required, no charge if you cancel during the trial.