Quick Insight
Cross-shopping analysis measures the overlap in customers between two or more brands, locations, or categories within a defined time window, usually expressed as the share of one place’s visitors or buyers who also visited or bought at another. It is built from aggregated, anonymized behavioral data, either device-level visits or card transactions, matched to a reference set of points of interest. Unlike market share, which compares total sales, it shows how the same customers split their trips or spend across competitors.
How Cross-Shopping Analysis Works
- Build the place layer: Every store needs a consistent name, brand ID, category, and location. A clean POI dataset makes “Target” in one source match “Target #1123” in another.
- Attribute activity to places: Device pings are matched to store boundaries, or card transactions are matched to merchant locations. Building footprint polygons cut errors in dense spots like strip malls, where a radius around a pin can assign a coffee shop visit to the dry cleaner next door. This step is the core of visit attribution.
- Choose a time window: Same trip, same day, or same month. Short windows surface complementary stops; long windows surface brand switching.
- Calculate overlap: Find the share of Store A’s customers who also went to Store B. Overlap is asymmetric: 40% of a small boutique’s shoppers might visit the nearby Walmart, while only 1% of that Walmart’s shoppers visit the boutique.
- Normalize: An affinity index compares observed overlap with what each brand’s size alone would predict, so giant chains don’t top every list. Scales differ by platform.
Types of Cross-Shopping Analysis
- Visit-based: Uses mobile location data. Captures physical trips, including browsing without buying.
- Transaction-based: Uses card spend data. Captures actual purchases, including online ones, but misses cash and visits with no purchase.
- Session-based: Uses web browsing data, such as shoppers who view two brands within one 30-minute session. Digital only.
Why Cross-Shopping Analysis Matters
- Reveals real competitors: The brand a team assumes it competes with is often not the one sharing the most customers.
- Separates convenience from choice: A shopper who fills up at the gas station beside their grocery store is proximity overlap. A shopper who drives past one grocer to reach another is intentional overlap, and it tells a very different story.
- Adds context to market share: Two brands can hold flat share while customers quietly shift between them.
Common Uses
- Competitive benchmarking: Track which rivals are gaining shared customers over time as part of a wider location intelligence program.
- Site selection and co-tenancy: Open near brands whose shoppers already overlap with yours.
- Shopping center leasing: Shape tenant mix and placement around tenants that share foot traffic.
- Conquest and partner marketing: Target a competitor’s shoppers, or test a co-promotion with a complementary brand before spending on it.
- Investment research: Watch brand switching as an early signal ahead of reported earnings.
Pros and Cons
Pros | Cons |
Based on observed behavior, not survey answers | Device and card panels can skew toward certain age or income groups |
Works at brand, category, and single-store level | Low-traffic stores may fall below privacy thresholds and go unreported |
Repeatable, so trends are easy to track | Only as accurate as the place matching underneath it |
Covers competitors you have no first-party data on | Shows that overlap exists, not why |
Want to learn more about how competitive intelligence works?
Read our full guide: The Ultimate Guide to Competitive Intelligence Research
Frequently Asked Questions
1. What counts as a “high” overlap?
It depends on the category. Grocery and big-box overlap runs high almost everywhere, so a 30% overlap there means less than 30% between two specialty retailers. Compare against category averages, not a fixed number.
2. How often should cross-shopping data be refreshed?
Monthly is the common cadence. Shorter periods get noisy, while quarterly can miss a competitor’s new opening or a promotion’s effect.
3. Can cross-shopping analysis identify individual shoppers?
No. Reputable providers report only aggregated percentages, so you see that 22% of a store’s visitors also went elsewhere, never who they were.
4. Can I run it using only my own loyalty or CRM data?
Only partly. Your own data shows overlap between your own stores or product lines, but it can’t show where those customers go when they leave you.
5. Does it work outside retail?
Yes. The same method applies to gyms, restaurants, hotels, healthcare clinics, and entertainment venues: any place where people choose between options.
