Top of funnel
Paid social measured past the platform's own report.
Paid social prospecting measured past the platform's own view-through count, using geo holdouts and matched-market tests to show what campaigns actually produce.
This page covers prospecting — campaigns built to reach people who do not know the business yet. Re-engaging people who have already visited a site or interacted with a page is a different discipline with a different audience, a different message, and a different economics, covered on the retargeting page. Prospecting competes for attention from a cold audience; retargeting works a warm one. Measuring both the same way would hide how differently they actually perform.
Meta will report more conversions from a campaign than actually happened, and the mechanism is not fraud — it is view-through attribution. The platform counts a conversion any time someone shown an ad later converts within its attribution window, whether or not that person ever consciously noticed the ad, clicked it, or was influenced by it at all. An impression scrolled past in half a second gets treated the same as a deliberate click. That is why the number in Ads Manager is reliably higher than the number in a CRM, and why comparing the two directly makes paid social look worse than it is, or better than it is, depending on which side of the gap gets quoted.
We resolve that by testing prospecting's actual effect directly — geographic holdouts and matched-market tests that show what happens to bookings in markets where a campaign runs against markets where it does not.
How the work differs
- Creative volume
- A steady cadence of new creative treated as an ongoing dependency the program runs on, since prospecting fatigues an audience faster than retargeting does.
- Geo holdout testing
- Comparable markets held back from a campaign on a rotating basis, so the lift it produces can be measured against a real control rather than assumed from a before-and-after comparison.
- Matched-market testing
- Where a true geographic holdout is not practical, markets matched on size, demographics, and historical performance are compared instead, with a difference attributed to the campaign only once it clears a meaningful threshold.
- View-through discounting
- Platform-reported view-through conversions logged and reported on their own separate line from click-through and matched outcomes.
- Audience structure
- Prospecting audiences built to exclude existing customers and recent site visitors, since including them would let retargeting's effect masquerade as prospecting's.
- Platform allocation
- Budget split across Meta, LinkedIn, or TikTok based on where the target audience actually spends attention for this specific business.
How it is measured
The number reported is lift, measured through a holdout or matched-market comparison and expressed as bookings attributable to a campaign running versus not running, not the platform's own reported conversion count. That lift figure is what feeds into the attribution layer, reported on the same terms as every other channel's number.
What this channel cannot claim credit for: a conversion from someone who had already decided to buy and would have converted through another channel regardless of the ad. Prospecting campaigns are measured on incremental lift specifically to avoid that overlap; a platform's raw conversion count makes no attempt to separate the two.
What we need from you
A steady supply of creative — at minimum a handful of new concepts monthly, since audience fatigue sets in faster on prospecting campaigns than most teams expect. Administrative access to the ad accounts across whichever platforms are in use. And agreement on which markets can be held back for testing purposes, since a genuine holdout requires leaving real demand unaddressed in a comparison market for the length of the test.
Who this is for
Built for operators who want to know what prospecting is actually producing before deciding how much of it to run, especially multi-location businesses where holding back a comparison market is operationally possible in a way it rarely is for a single-location business. Which platform gets the budget differs by audience — a storage portfolio and a healthcare group rarely reach the same people in the same place, and campaign structure follows that rather than a fixed template.
This is one of nine top-of-funnel programs. See how an engagement runs for the onboarding timeline, or client results for a finished report.
Because Meta counts view-through conversions — someone shown an ad who converts later without ever clicking it or, in many cases, consciously noticing it — inside its own attribution window, and that number is always going to be larger than a count based on people who actually engaged with the ad or were in a market where the campaign measurably moved bookings. It is not a reporting error on either side; the two numbers answer different questions. Our number answers whether the campaign caused incremental bookings, which is what a budget decision should actually be made against. Meta's number answers something closer to reach, which has its own value but is not the same thing.
Enough that the comparison markets have meaningful volume to compare — there is no fixed number, since it depends on how many comparable markets exist and how much natural variation those markets already show month to month. A minimum is calculated before recommending a holdout, the same way a sample size is calculated for a conversion test, and we will say directly if current spend is too thin to produce a readable result yet. Running a holdout at too small a scale produces noise dressed up as an answer, which is worse than not testing at all.
It depends on spend level and audience size, but a functioning prospecting program typically needs new creative concepts monthly at minimum, not quarterly, since audience fatigue on prospecting campaigns sets in faster than most teams budget for. Treating creative as a fixed asset produced once and run indefinitely is one of the most common reasons a paid social program's cost per result climbs steadily over a quarter. The creative cadence is sized to the spend level, and we flag when volume is falling behind what the program needs to stay effective.
They are rarely the same. A healthcare group's prospecting audience often responds better to platforms where longer-form trust-building content performs, while a self-storage portfolio's audience frequently responds to more immediate, visual, location-specific creative on a different platform mix. We do not start from a fixed platform allocation and adjust it slightly by industry; the allocation is built from where the specific target audience actually spends attention, tested rather than assumed.
See what your prospecting campaigns actually produce.
A holdout or matched-market test that separates real incremental bookings from view-through conversions the platform is already counting on its own.