Foot Traffic Measurement 101: Connecting Ad Spend to Store Visits - InMarket

Every marketer has sat through the meeting where leadership asks the same question a different way: did this actually work? Impressions, clicks, and reach can show that a campaign ran and that people saw it. None of those numbers can tell you whether it changed anyone’s behavior.

That gap is becoming harder to paper over. According to the IAB’s State of Data 2026 report, 75% of US buy-side leaders say their core ad measurement approaches — attribution, incrementality testing, and marketing mix models alike — are underperforming. For any brand where the real-world outcome is a physical visit — retail, QSR, auto, restaurants — the question leadership actually wants answered is simpler and harder: did this media get someone in the door who wouldn’t have shown up anyway?

Answering that requires a different discipline than attribution. It requires incrementality.

Attribution vs. incrementality: not the same question

Attribution asks: who was exposed to an ad, and did they later visit or buy? It’s observational. It counts people who showed up after seeing a campaign — including the ones who were always going to show up. A regular at a QSR chain who sees an ad and then, unsurprisingly, visits that week gets counted as a win, even though the ad changed nothing.

Incrementality asks a different question: how many additional visits happened because of the campaign that would not have happened otherwise? It’s a counterfactual exercise — modeling what visitation would have looked like in a world where the campaign never ran, then measuring the gap between that prediction and what actually happened. That gap is the incremental lift. Everything else is noise: seasonality, existing demand, habit, a competitor closing down the street.

The distinction matters because attribution and incrementality can tell opposite stories about the same campaign. A channel can show strong attributed visits and zero incremental lift — it was simply reaching people who were already coming in.

What Visit Lift measurement actually does

Visit Lift measures the incremental increase in physical store visits attributable to a media campaign — not total visits, but the extra visits the campaign caused. At a high level, without getting into proprietary mechanics, the approach works like this:

Identify who was exposed to the campaign across the channels being measured — the tracking layer that flags exposure across mobile, TV, streaming audio, OOH, and more.
Build a statistical model of expected visitation based on historical patterns — a synthetic control that estimates what visits would have looked like without the campaign.
Compare actual visits during the campaign to that modeled expectation.
The difference is incremental lift — the visits the media can actually take credit for.

This matters most for categories where a store visit is the KPI: retail, QSR, auto, restaurants. For these brands, a defensible visit lift number is closer to the bottom line than a click ever was.

A national dining chain saw this play out directly. Running a social campaign to launch a new seasonal menu item, the brand measured a 26% incremental visit lift versus a 23% benchmark — 285,000 incremental visits attributable to the media, not just visits that happened to follow it. Breaking the results down by ad frequency and channel placement showed even wider variation (a 31% lift at 1–5 ad exposures, a 29% lift tied to specific channel placement), which is exactly the kind of dimension-level detail a simple exposure-match can’t surface.

Why a counterfactual approach beats simple matching

Not all “foot traffic measurement” is built the same, and it’s worth being skeptical of anything that claims otherwise without explaining its method. Some approaches simply match observed visits to ad exposure: who was exposed, and who later showed up? That’s a correlation, not a causal claim, and it tends to overcount, which is the same problem attribution has.

A causal, counterfactual approach asks a more rigorous question: build a proper baseline first, then isolate the marketing effect from natural demand, seasonality, and existing trends. Think of it the way a clinical trial works — you don’t just measure whether patients who took a drug got better. You compare them to a control group that didn’t, because some of them would have gotten better anyway. Marketing measurement has the same problem, and the same fix: a control.

What this unlocks for marketers

Getting incrementality right changes what marketers can actually do with the data:

Defensible conversations with finance. ROI grounded in a real-world outcome — a store visit — holds up in a budget review in a way proxy metrics don’t.
In-flight optimization. Instead of waiting for a post-campaign report, budget allocation can shift mid-flight toward whatever channel is proven to be driving incremental visits.
Cross-channel comparability. Visit lift can be measured across TV, streaming audio, OOH/DOOH, display, social, and more, putting channels that are usually measured in completely different ways on the same footing.
Deeper diagnostic detail. Which audiences, creative, markets, and frequencies are actually driving visits — useful for both optimizing the current campaign and planning the next one.

A casual dining brand’s holiday campaign shows why that diagnostic layer matters. A seasonal beverage promotion produced a 10% incremental visit lift, and a 17% incremental sales lift (against an 11% benchmark), totaling $197K in incremental sales. Breaking it down by creative, the “holiday interest” variant alone drove an 11% incremental visit lift — the kind of dimension-level read that tells a brand which creative to lean into next time, not just whether the campaign worked overall. It’s also a useful example of the sales-and-visits point below: the same measurement approach captured both outcomes in one read, rather than treating them as separate stories.

Three misconceptions worth retiring

“More impressions means more visits.” Not past a point. Ad frequency has diminishing returns, and saturation is real — at some threshold, additional impressions stop producing additional incremental visits.

“If someone visited after seeing my ad, my ad worked.” This is the attribution trap again. A visit that would have happened regardless isn’t evidence the campaign worked — it’s evidence the campaign ran.

“Foot traffic measurement doesn’t matter if I sell online too.” Increasingly, brands and modern measurement approaches don’t treat online and in-store as separate stories. Media drives both kinds of outcomes, and unifying sales and visit measurement gives a more complete read on what’s working.

Where this is headed

Privacy shifts like identity changes, opt-in requirements, and an increasingly uncertain future for third-party signals are pushing the entire measurement industry toward approaches that are first-party, permission-based, and statistically rigorous by design. Simple exposure-matching doesn’t hold up well in that environment. Counterfactual, incrementality-based methods are built for it.

As the identity landscape keeps shifting, the brands with a defensible, causal measurement approach in place now will be the ones still able to answer “did this work?” with a straight answer later.

See how Outcome Intelligence powers InMarket’s approach to incrementality and Visit Lift measurement. Get in touch at InMarket.com/Contact.