Practical framework / 5 MIN READ
Keep store visits separate from store sales
Use modeled footfall as a defined signal with eligibility and causal limits.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-06
An advertising report may estimate visits associated with prior ad interactions. That is different from counting every person entering the store, verifying a purchase or proving the visit would not otherwise have happened.
What you will leave with: Prepare a physical-location measurement brief that keeps modeled visits, operational footfall and purchases distinct.
Use this when
- A retailer wants to attach a sales value to every reported store visit.
- A store-visit metric disappears or changes after reporting updates.
This is not permission to track individual customers’ locations or a method for proving offline incrementality.
Define the location outcome
Write the decision the business needs to make. Increasing suitable store traffic, increasing profitable purchases and improving visit experience are related but different goals. A visit can be useful without a purchase, while a store can become busier without earning more.
Map available aggregate evidence: point-of-sale transactions, door counts, appointment records and platform visit estimates. Each has its own unit and coverage. Do not force them into a single total until those definitions are understood.
Read the model and eligibility conditions
Google describes store visits as modeled numbers based on aggregated information and specifies eligibility requirements. Treat the resulting measure as a platform estimate, not a list of identified visitors. Availability can depend on adequate data and supported locations.
Confirm that the business locations and report settings are correct. If reporting is unavailable, do not interpret the blank as zero visits. Avoid efforts to reconstruct identities or infer sensitive location visits. The limits are part of the measurement design, not obstacles to bypass.
Assign value only with defensible assumptions
If planning needs an estimated value per visit, state how purchase likelihood and relevant order economics are obtained. Use comparable store and customer context, and separate observed revenue from the modeled value. A single average can be misleading when stores or visits differ substantially.
Do not add modeled store value to online revenue without examining overlapping journeys and the meaning of each value. A customer might research online, visit a store and later buy through the website. The same commercial outcome should not become several independent sales merely because it appears in several reports.
Choose the supported conclusion
Use visit estimates for the descriptive or optimization purpose they can support, with their limitations stated. A campaign-associated visit is not automatically an incremental visit. For a causal question, commission a suitable evaluation with appropriate safeguards and specialist design.
Reconcile broad trends with operating context: opening hours, store closures, promotions and stock availability. Preserve reporting maturity and configuration changes. A useful readout says what changed in each measure and what remains unresolved about the relationship between advertising, visits and purchases.
Make the reasoning concrete.
A hypothetical retailer sees 200 modeled store visits in an ad report and 80 point-of-sale purchases during the same period. It proposes multiplying the 200 visits by the average receipt and calling the result advertising revenue.
Value every visit as a purchase
Opening: Apply the full average receipt to each estimated visit.
Hypothesis: Some visitors may have purchased.
Risk: Visits and purchases are different outcomes, and causality is unproven.
Keep measures separate
Opening: Report modeled visits and actual sales with an explicit valuation scenario if needed.
Hypothesis: The team can inspect assumptions without confusing them with revenue.
Risk: Aggregate totals do not reveal which purchases followed which visits.
Do not report the multiplication as measured sales. Use a clearly labeled planning assumption and retain the distinction between association and incrementality.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Decision
Is the goal traffic, purchases or profitable sales?
Coverage
Which locations and visit estimates are eligible?
Valuation
What evidence supports purchase probability and value?
Overlap
Could online and offline figures describe the same outcome?
Claim
What would a causal conclusion require?
Watch for the wrong conclusion.
- Treating modeled visits as identified people.
- Reading unavailable reporting as zero visits.
- Valuing every visit as a completed sale.
- Adding overlapping online and offline value without reconciliation.
Your next action
Write a store-measurement glossary separating visits, transactions and incremental outcomes before presenting an omnichannel total.
Sources & scope
Frameworks, worksheets and hypothetical examples are original editorial synthesis. Linked primary sources support only the stated concepts. Source checking is not an independent audit of effectiveness. No named human author or reviewer is claimed.
- Google Ads Help: About store visit conversions →Primary platform documentation · Supports: Describes modeled, aggregate store-visit reporting and eligibility requirements. Store visits are not verified purchases or an experimental estimate of incremental visits.