Can We Measure the Impact of DV360 on Store Visits and Purchases?

The short answer: yes, but not with a single number

One of the most common questions from marketing leaders evaluating Display & Video 360 (DV360) is whether digital media can be tied back to what happens in the real world — footfall in stores and purchases at the till or online. The honest answer is that DV360 can measure impact on store visits and purchases, but doing it credibly means combining the platform's native measurement tools with sound experimental design.

This guide breaks down the measurement options available, what each one can and can't tell you, and how to move from correlation ("sales went up") to genuine incrementality ("sales went up because of DV360").

What DV360 can measure natively

DV360 sits inside the Google Marketing Platform (GMP), which gives it access to several outcome signals beyond clicks and impressions.

Online conversions and purchases

The foundation is conversion tracking. Through Campaign Manager 360 (CM360) Floodlight tags, or a Google Analytics 4 integration, DV360 can attribute:

• Online purchases and revenue
• Add-to-cart, lead form and sign-up events
• Assisted conversions across the path

This is the most reliable layer because the conversion event is captured directly on your own properties. With enhanced conversions and server-side tagging becoming standard, this measurement is also more resilient to browser signal loss.

Store visits

Store visit measurement estimates how many people who saw or interacted with your ads later visited a physical location. It uses aggregated, anonymised, and modelled location data from users who have opted in to location history. Because it is modelled and privacy-safe, store visits are reported as estimates at the campaign or line-item level — never as individually identifiable people.

Availability depends on meeting Google's eligibility thresholds: a minimum number of store locations, sufficient ad volume, and adequate modelled data density in your markets. Not every advertiser or country qualifies, so it's worth confirming eligibility before you build a strategy around it.

Offline sales via data uploads

Where purchases happen offline — in-store, over the phone, or through a sales team — you can bring that data back into the ecosystem:

• Offline conversion imports into CM360/GA4 tie CRM or point-of-sale records back to ad exposure.
• Customer Match and first-party data let you connect known customers to campaign activity within privacy rules.

This closes the loop between a DV360 impression and a transaction that never touched a website.

The measurement layers, side by side

Method	What it measures	Data source	Best for
Floodlight / GA4 conversions	Online purchases, leads, events	Your own site/app	E-commerce and lead gen
Store visits	Modelled physical footfall	Aggregated location data	Retail, QSR, automotive
Offline conversion import	In-store/CRM sales	Your CRM or POS	Considered purchases, B2B
Conversion lift / Brand lift	Incremental effect	Randomised experiment	Proving causality

Attribution tells you where — experiments tell you whether

Attribution models (last click, data-driven, position-based) distribute credit across touchpoints. They're useful for understanding the customer journey and reallocating budget, but they share a fundamental limitation: they only measure people who converted, not whether your ads caused the conversion.

Someone who was going to buy anyway, and happened to see your ad, still gets counted. To separate real impact from coincidence, you need experimentation.

Conversion lift and store visit lift

Google offers randomised controlled experiments where a portion of your addressable audience is held out from seeing ads. The platform then compares conversion — or store visit — rates between the exposed and control groups. The difference is your incremental lift: the outcomes that would not have happened without the campaign.

This is the gold standard for answering "did DV360 actually drive purchases?" rather than "how many purchasers saw an ad?"

Geo experiments

Where user-level holdouts aren't practical, geo-based testing splits regions into test and control markets. You run DV360 in some geographies and suppress it in comparable ones, then measure the difference in store visits or sales. Geo tests are especially valuable for omnichannel retailers because they capture total business impact — online and offline combined — without relying on cookies or device IDs.

How to build a credible measurement plan

Measuring real-world impact well is a process, not a toggle. A practical sequence looks like this:

1. Define the outcome that matters. Store visits, revenue, new customers — pick the metric your business is actually judged on, not a proxy.
2. Instrument the plumbing. Confirm Floodlight or GA4, server-side tagging, and offline import feeds are all firing accurately before you spend at scale.
3. Check store visit eligibility. Validate location counts, volume and market coverage early.
4. Run a lift or geo test. Bake incrementality measurement into flagship campaigns rather than bolting it on afterwards.
5. Reconcile against business data. Compare platform-reported outcomes with your finance or POS numbers to build internal trust.

Getting this foundation right is where a lot of teams struggle, and it's a core part of what our DV360 managed services and co-managed services engagements focus on.

Common pitfalls to avoid

• Double counting. Store visits, online conversions and view-through credit can overlap. Agree definitions before reporting to stakeholders.
• Treating modelled data as exact. Store visits are directional estimates. Use them for trend and comparison, not to the decimal point.
• Ignoring view-through settings. Overly generous view-through windows inflate apparent impact. Align windows to realistic purchase cycles.
• Skipping the control group. Without a holdout, you're reporting activity, not impact.
• Signal loss blind spots. As third-party identifiers decline, lean on first-party data and modelled, aggregate methods rather than fighting the trend.

What good looks like

A mature DV360 measurement setup produces a layered view:

• Real-time conversion and revenue reporting for optimisation.
• Modelled store visit trends for offline-heavy businesses.
• Periodic lift or geo experiments that quantify incrementality.
• A reconciliation habit that ties platform numbers to actual sales.

Together these let you answer the CFO's question — "what did this spend return?" — with evidence rather than assertion. If you're setting up your account or reviewing your measurement stack, our DV360 self-serve and full services overview explain the options for different team structures.

Bringing it together

DV360 absolutely can measure its impact on store visits and purchases — through native conversion tracking, modelled store visits, offline data imports, and, most importantly, randomised and geo experiments that prove causality. The platform provides the signals; the rigour comes from how you design measurement around your specific business outcomes.

If you want help building a measurement framework that connects DV360 spend to real footfall and revenue — and stands up to scrutiny from your finance team — talk to our DV360 experts. We'll help you instrument the tracking, design the right lift or geo test, and turn platform metrics into decisions you can defend.