Skip to content

Cross-Shop & Basket Affinity Analyzer

Map which brands and categories your shoppers also buy over time.

What is the Cross-Shop & Basket Affinity Analyzer?

The Cross-Shop & Basket Affinity Analyzer is a free AI skill that reads a shopper base's broader buying relationships across trips and time, not just same-trip basket pairs. You give it your brand, the loyalty or panel data you already have, and the category or timeframe you want to examine; it returns a ranked map of the other brands and categories your shoppers also buy over that period, a household-level overlap read, the strategic implications — co-marketing candidates, bundle ideas, or media targeting — and the assumptions to validate. It is built for shopper insights and marketing managers who already have a same-trip basket analysis and need the next layer: what a shopper's relationship to your brand looks like across their whole shopping life. Because it works from cross-trip purchase history instead of a single transaction, it surfaces opportunities a basket-pair analysis cannot see. Layering in live food and beverage demand data shows whether a cross-shop pattern reflects a rising relationship or one that's already fading.

Who it's for

  • Shopper insights managers analyzing loyalty or panel data
  • CRM and loyalty analytics leads scoping co-marketing partners
  • Category management partners studying household shopping relationships
  • Brand strategy teams targeting media based on cross-shop behavior

What you get

  • A ranked cross-shop map of brands and categories
  • A household-level overlap read
  • Co-marketing and bundling candidates surfaced by the pattern
  • Media and targeting implications
  • A narrative summary for a stakeholder audience
  • Assumptions and data-quality caveats to validate before acting

How to use it

  1. 1Paste the prompt into ChatGPT, Claude or Gemini.
  2. 2Fill the bracketed inputs (brand, data you have, category, timeframe).
  3. 3Answer its clarifying questions about your data's coverage and limits.
  4. 4Ask it to shortlist the top 3 cross-shop candidates for a co-marketing pitch.

See it in action

Example input

Brand: fictional 'Halcyon Oat Milk'. Data: 90-day loyalty-card purchase history for about 22,000 Halcyon buyers at a regional grocery chain. We can see that 38% of Halcyon buyers also bought a plant-based creamer, 29% bought cold brew coffee, and 24% bought granola in the same window. Goal: find a co-marketing partner category for a spring campaign.

Example output

Cross-shop affinity read — Halcyon Oat Milk, 90-day window

Data recap
Base: ~22,000 Halcyon Oat Milk buyers, 90-day loyalty history, one regional grocery chain. Observed cross-shop rates: plant-based creamer 38%, cold brew coffee 29%, granola 24%. Treat these three as your strongest observed signals; anything below this in your data is likely a longer tail worth a second look but not the lead candidate for a single spring campaign.

Ranked cross-shop categories

1. Plant-based creamer (38% overlap)
Read: the highest overlap, and the most adjacent use-case — a shopper already comfortable with plant-based dairy alternatives for one occasion (drinking) is highly likely to use one for another (coffee whitening). This is likely more habit-and-identity driven than occasion-driven, which makes it a strong but less "campaign-able" pairing — these shoppers may already consider themselves a plant-based household broadly, rather than being newly recruited by a pairing message.

2. Cold brew coffee (29% overlap)
Read: a genuine occasion pairing — oat milk and cold brew share a morning-routine moment. This is the strongest candidate for a spring co-marketing campaign specifically because it is occasion-based rather than identity-based: a "cold brew + oat milk morning" message can plausibly recruit trial in both directions, not just describe an existing habit.

3. Granola (24% overlap)
Read: a breakfast-occasion adjacency, weaker than cold brew because the day-part fit is looser (granola spans breakfast and snacking) and the overlap rate is the lowest of the three. Worth a secondary mention in bundle creative but not the lead partner category.

Household-level overlap notes
A 38% overlap on plant-based creamer suggests just over a third of your Halcyon base is buying into a broader plant-based-dairy pattern, not a single-product habit — useful context for messaging beyond this campaign, since it implies category-level loyalty rather than brand-specific loyalty alone. Confirm whether your data can also show whether these are the SAME plant-based creamer brand repeatedly or a mix, since that changes whether a specific creamer brand is a strong single co-marketing partner or whether the opportunity is more "plant-based creamer as a category."

Strategic implications
- Lead candidate: a cold brew coffee brand or category co-marketing partnership for the spring campaign — the occasion fit is real and the overlap is strong enough to act on.
- Bundle idea: a "cold brew + oat milk" in-store or digital pairing, timed to when cold brew sales typically pick up seasonally.
- Media targeting: if your media platform supports interest-based or purchase-based targeting, cold brew and coffee-adjacent audiences are a reasonable lookalike pool for Halcyon media, given the observed overlap.

Assumptions to validate
- These are correlational overlaps from one 90-day window at one retailer; confirm the pattern holds across a second time window or retailer before committing meaningful co-marketing budget.
- The direction of the relationship (oat milk drives cold brew purchase, or vice versa, or both driven by a third factor like a broader wellness routine) is not established by this data alone.

Want me to draft the co-marketing pitch angle for a cold brew partner, or run this same analysis for a second time window to confirm the pattern holds?

The prompt

Here's the start of the prompt. Download the free bundle for the full, ready-to-paste version — plus the installable Claude Skill and Custom-GPT instructions.

# Role
You are a senior shopper insights analyst who studies cross-shop behavior — what a shopper base buys across other brands and categories over time — as distinct from single-trip basket analysis. You separate genuine occasion-based pairings from identity-driven habits, and you never overstate what correlational data proves.

# Context I'll provide
- Brand: [BRAND]
- Data available: [DATA — e.g. loyalty-card history, panel data, what timeframe and sample size]
- Category or brands observed in the cross-shop data: [OBSERVED CATEGORIES + overlap rates or notes, if known]
- Timeframe: [TIMEFRAME]
- Goal: [GOAL — e.g. find a co-marketing partner, sharpen media targeting, understand household behavior]

# Your task
1. If the data source, timeframe, or goal are missing or vague, ask up to 3 clarifying questions BEFORE writing anything.

Frequently asked questions

What is cross-shop and basket affinity analysis?
Cross-shop and basket affinity analysis looks at the other brands and categories a shopper base buys across multiple trips and a defined time window — not just what lands in the same basket on one visit. It reveals broader shopping relationships, like whether your buyers are also a plant-based-dairy household or a cold-brew-coffee household, which single-trip basket data can't show. This skill turns loyalty or panel data into a ranked map of those relationships and what to do with them.
How is this different from the Market Basket & Co-Purchase Analysis Narrator skill?
The Market Basket & Co-Purchase Analysis Narrator looks within a single trip — which items shoppers put in the same basket together, like chips and salsa. This skill looks across trips and time — which other brands and categories a shopper base buys over weeks or months, even on separate visits. They answer different questions: same-trip pairing supports a display or bundle; cross-trip affinity supports co-marketing partnerships and media targeting based on a household's broader shopping identity.
What AI models can run this prompt?
Any capable chat model — ChatGPT, Claude, or Google Gemini. The prompt is model-agnostic, so paste your loyalty or panel data summary directly into a chat, or save the prompt as a Custom GPT or reusable skill for recurring cross-shop reads across different brands or time windows.
What kind of data do I need to run this?
Loyalty-card or panel data showing what else your branded buyers purchased over a defined window works best, even as a simple summary of top overlapping categories and their rates. The skill will not invent overlap percentages or sample sizes on your behalf — bring real numbers, even rough ones, and it will flag thin data or a single time window as a limitation rather than presenting the pattern as settled fact.

Related skills

Want the live data behind sharper outputs?

These skills get better with real-time F&B intelligence. See what Tastewise can do for your team.