Comparing two brands

When you want to understand how two brands’ archetype neighborhoods differ geographically, select a second brand from the brand dropdown. The map and rankings shift to reveal where one brand’s neighborhood style dominates.

The diverging map

When comparing two brands, the map switches from a single-color similarity scale to a diverging lean map:

• Hue shows which brand a neighborhood leans toward, using a fixed, colorblind-safe blue-and-orange pair (blue for Brand A, orange for Brand B) rather than the brands’ own colors. This is deliberate: many natural pairings would otherwise be red-versus-green, which about 8% of men cannot distinguish, and the lean direction is the whole point of this map.
• Brightness shows how strongly that neighborhood resembles the more-similar brand (0–100 scale).
• Median tracts stay dark on purpose — they resemble both brands roughly equally and don’t light up.

This visual design highlights neighborhoods that are geographically distinctive between the two brands, so you can spot where one archetype clusters and the other does not.

Note

Percentiles in a compare view are within-brand ranks (Brand A scores are how that neighborhood ranks among Brand A’s potential fit; Brand B scores are how it ranks among Brand B’s potential fit). These percentiles are not comparable as absolute strength across brands. A tract at 85th percentile for Brand A and 75th percentile for Brand B may actually resemble Brand A much more strongly, or the opposite — the percentiles reflect each brand’s internal distribution. Always read the side-by-side percentiles and the lean drivers (below) to understand the actual difference.

The three rankings

Below the map, three buttons let you filter neighborhoods into categories:

• Leans [Brand A] — neighborhoods that score notably higher for Brand A. Only shown if the difference is meaningful (minimum 60th percentile for Brand A).
• Resembles both — neighborhoods that score at or above 80th percentile for both brands. These are rare; the more distinctive two brands are, the fewer neighborhoods fit both equally well.
• Leans [Brand B] — neighborhoods that score notably higher for Brand B.

Each list ranks neighborhoods within its category. The count under the heading shows how many tracts in the current scope (nation-wide or a selected metro) meet the “Resembles both” threshold.

Selected neighborhood detail

Click any neighborhood in a ranking to see its side-by-side profile:

• Percentile bars for each brand show the tract’s similarity score to each archetype.
• Nearest store shows the distance (in kilometers) to the closest store of each brand.
• Why the difference — a plain-English summary of which features drive the lean. For example: “It leans Whole Foods-ward (by 12 percentile points), mainly on median home value and bachelor’s degree attainment.”

The panel also shows a “Leans both equally” message if the two percentiles are within 4 points of each other.

What the lean shows

A high lean toward Brand A does not mean Brand A will succeed there, nor that Brand B will fail. It means that neighborhood’s observable characteristics (income, density, education, retail density, age structure, ethnicity, transportation) are closer to the average of Brand A’s current locations than to Brand B’s.

Use the lean as a discovery signal: “Where does Brand A’s neighborhood type naturally recur, independent of Brand B?” or “Which neighborhoods look different to these two brands, and why?”

Remember: resemblance is not prediction

A high lean for Brand A reflects neighborhood similarity, not sales potential. Brand success depends on many unmeasured factors: brand awareness, competition, real estate availability, store operation, supply chain, pricing, and consumer behavior. This tool helps you triage neighborhoods worth investigating further, not forecast revenue or foot traffic.

How the lean is computed

Both brands are scored on the same scoring method. For each neighborhood:

1. A similarity percentile is computed for Brand A (0–100, within Brand A’s distribution).
2. A similarity percentile is computed for Brand B (0–100, within Brand B’s distribution).
3. The lean is the difference: Brand B percentile minus Brand A percentile.
4. Neighborhoods are ranked by this difference (within their filter category).

The method page explains how features are weighted, how archetype coherence is validated, and why percentiles are brand-specific. The lean value itself should be read as a directional signal, not an absolute strength measure — a 20-point lean reflects a material difference in the feature mix, but does not map to any standard measure of market suitability.

Combining with factor weights

The compare view respects any custom factor weights you set. If you reweight factors while comparing, the map and rankings recalculate immediately, and leans may shift. This lets you ask: “How would the lean change if we weight Urban density more heavily?” A brand’s lean toward urban neighborhoods may shrink or vanish if you downweight density, revealing how dependent the difference is on that one factor.

Tip

If two brands’ leans seem unstable across weight presets, it means their neighborhood styles are similar on the dimensions you’re emphasizing. A stable, consistent lean across presets suggests the two brands have genuinely different archetype profiles.

Comparing a custom upload against a brand

You can also compare your own uploaded footprint against a built-in brand. Upload your store list, then pick a built-in brand to compare against. Side A is your upload’s plain resemblance percentiles (not the avoid-failures contrast score used in the ranking, which is not a symmetric “how similar” measure), and side B is the built-in brand. This answers the headline expansion question directly: “how does my footprint’s resemblance surface differ from Whole Foods’?” The same not-comparable-across-brands caveat on the percentiles applies.

Next steps

• Dive into a metro: Use metro focus to rank neighborhoods within a specific city or region, then compare how the two brands’ archetypes play out locally.
• Explore one brand alone: Exit the compare view by clicking the × button, and focus on a single brand’s full neighborhood profile.
• Upload your own stores: If you’re doing a competitive analysis with your own retail locations, you can upload your store list and compare your brand footprint to an industry benchmark brand.
• Export findings: Use the shortlist and reporting feature to save and annotate neighborhoods of interest, then export a PDF or CSV.