Marketing
·
6 min read
Your campaign probably worked better than your dashboard says
Ad platforms report online conversions. Most of the impact shows up in your stores, where no ad platform can see it. Here's how to measure what your marketing actually drives.
Avia Research · Retailogists
If you run marketing for a retailer with physical stores, you’ve probably seen this: a campaign drives strong reach, the store team says weekends were busy, but the ad dashboard shows a disappointing return. The budget gets cut.
The campaign may have worked. You just measured the wrong channel.
Most of the lift happens offline
One of the best-known field experiments on this question ran display ads on Yahoo! for a major US retailer with both stores and a website (Lewis & Reiley, “Online Ads and Offline Sales”). The researchers matched ad exposure to the purchase records of about 1.6 million customers and compared them with a randomised control group.
The results:
Customers who saw the ads bought about 5% more than the control group.
93% of that lift happened in brick-and-mortar stores, not online.
The extra revenue was more than 325% of what the ads cost.
An online-only report would have credited the campaign with about 7% of what it actually drove.
The gap hasn’t closed with better tracking. A 2026 analysis by LiftLab and PMG across enterprise omnichannel brands found that branded search’s true causal contribution was 30 to 70% lower than last-click reports suggested. Platforms over-credit the clicks they can see and miss the store visits they can’t.
Why store visits are the missing metric
For most retailers, the link between a campaign and a sale runs through a store visit. A shopper sees an ad, walks in a few days later, and buys at the counter. POS data captures the sale but not the cause. Ad platforms capture the impression but not the sale.
Foot traffic sits in the middle. It's the first measurable sign that marketing changed behaviour, and it lets you split campaign performance into two separate questions:
Did marketing bring more people in? Measured by traffic lift against a baseline.
Did the store convert them? Measured by conversion rate and revenue per visitor.
Keeping these apart matters. A campaign that lifts traffic 15% while conversion falls isn't a marketing failure. It's a staffing or stock problem in the store.
A simple method for measuring in-store lift
Set a baseline. Use at least four weeks of traffic by store, day of week and hour. Holiday weeks need last year’s pattern or a forecast, not a flat average.
Use control stores. Run the campaign in some markets and hold out comparable stores elsewhere. The difference in traffic change between the two groups is your lift.
Measure the full funnel. Track traffic, conversion and average transaction value for the campaign period and two weeks after, since store visits often lag exposure.
Convert lift into dollars. Incremental revenue = incremental visitors × conversion rate × average transaction value.
Compare with spend. Divide incremental revenue by campaign cost to get an in-store return you can compare across channels.
A worked example (illustrative numbers): a store averages 1,200 visitors a week. During a two-week campaign it averages 1,380 while control stores stay flat, giving 180 extra visitors a week. With a 22% conversion rate and an $85 average transaction, that’s 180 × 0.22 × $85 = $3,366 in incremental weekly revenue per store. Across 14 stores and two weeks, that’s about $94,000 that would never appear in an ad platform report.
Spending smarter, with better evidence
Once you can see which campaigns move traffic, and which don’t, budget decisions get much easier. You can cut the activations that don’t bring people in and put more behind the ones that do.
That’s the real value of in-store attribution. It doesn’t just justify budget. It shows you where to cut and where to double down, before the busiest quarter of the year.
Sources
KEEP READING
More articles
NEW
Operations
Your busiest holiday hours are where you'll lose the most sales
Holiday traffic is coming back to stores. The research says conversion falls just as the doors get busiest, and the fix starts in October, not December.
6 min read
Read more →
NEW
Analytics
Stores still take 83 cents of every retail dollar. Most can't see what happens inside them
E-commerce gets the analytics. Physical stores get the revenue. Here's what the latest data says about that gap, and why closing it is the biggest opportunity in retail right now.
6 min read
Read more →
Operations
The sale your POS never sees: what out-of-stocks really cost
An empty shelf doesn't show up as a loss in your sales report. It shows up as nothing at all. Here's how big that "nothing" is, and how to catch it before the holiday weekends.
5 min read
Read more →
See what Avia finds in your store.
Book a 20-minute demo and we’ll walk through your store’s traffic, conversion and revenue per visitor.