# LookAlike - site-selection discovery and triage for retail and commercial real estate ## What LookAlike is LookAlike is a location-intelligence tool that finds the neighborhoods that look like the places where a brand already wins. It profiles the neighborhoods around a brand's existing stores, ranks every neighborhood in the country by resemblance to that footprint, and explains feature by feature why each neighborhood made the list. LookAlike supports discovery and triage for expansion planning. It measures observable neighborhood resemblance using open data. It does not predict sales, revenue, foot traffic, or whether a specific store will succeed. The product makes a precise, supportable claim: these neighborhoods look like the ones where the brand already operates, so they may be worth investigating. ## Key facts - 83,008 neighborhoods scored across the continental United States. A neighborhood is a census tract. - 21 open-data features describe each neighborhood. - 383 metro markets are available for screening. - Every match is explained rather than presented as a black-box score. ## Market screening The Markets report rolls the national resemblance surface up to the metro level. It scores 383 markets for the selected brand and helps expansion teams answer the question they usually start with: which market should we examine first? Each metro scorecard includes: - Fit: how strongly the metro's neighborhoods resemble the brand's footprint. - Opportunity: the share of open, high-resemblance neighborhoods, normalized so large metros do not rank highly merely because they are large. - Headroom: open lookalikes per existing store, together with competitor presence. - A direct path from the metro scorecard into Explore, scoped to that market. ## Explore Explore provides a national ranked map of neighborhoods. Select a brand, whether it is the user's brand or another US chain, and the choropleth highlights neighborhoods that most resemble the places where that brand already operates. Only high-similarity areas are colored so the map directs attention to where to look first. Users can filter for open markets where the brand does not yet have a store, overlay existing stores, inspect the distance from a candidate to the nearest store, enforce distance from the brand's stores, and avoid tracked competitors. Scoring runs live in the browser, so changes to weights and filters re-score and re-rank neighborhoods immediately without a batch job. ## Compare Compare places two brands on one diverging map. It can compare a concept with a rival or compare a built-in brand with an uploaded footprint. Each comparison produces three rankings: neighborhoods that lean toward one brand, neighborhoods that resemble both, and neighborhoods that lean toward the other brand. A per-neighborhood breakdown explains the direction of the result and the factors behind it. An overlap count shows how contested the shared ground is. ## Your footprint Users can begin with curated national brands, add another US chain by name, or upload their own store list and rank the country against their actual footprint. An optional performance column can weight an archetype toward the best stores. Closed or underperforming locations can be flagged so LookAlike scores resemblance to winners minus resemblance to failures. Uploaded footprints become versioned datasets in a private workspace and can be loaded again without another upload. Users can also mark poor matches as "not a fit" and inspect the traits those rejected neighborhoods share. ## Explain Every neighborhood match includes a fingerprint that compares the neighborhood with the brand's archetype. It identifies the features driving the match, the features that diverge, and a plain-language assessment. Weak matches and mismatches remain as legible as strong matches. The explanation includes: - A similarity percentile against the whole country. - An absolute fit band of Strong, Moderate, or Weak, measured against the brand's own store neighborhoods. - A radar fingerprint comparing the neighborhood and brand archetype across all 21 features. - A robustness check showing whether the rank survives equal weighting and the removal of each factor theme, or depends heavily on one theme. - Low-reliability badges for noisy estimates, based on small populations or the source data's published margins of error. ## Tune Every ranking is driven by visible, adjustable weights. Six factor themes open into per-feature controls, and the map and rankings re-score live in the browser. Users can reweight or disable themes, use balanced, affluence-led, urbanism-led, and community-led presets, and save named scenarios. Race and ethnicity can be excluded with one click, and the application explains why a user may choose to exclude them. A configured view can be shared as a link. ## Deliverables LookAlike turns exploration into a defensible shortlist. Users can star neighborhoods into a brand-specific shortlist and annotate each candidate with the reason for considering the market, who to contact, and the next step. Available deliverables include: - A ranked shortlist with annotations. - A print-ready PDF report for a broker or franchise developer to share with a client. - A CSV export containing fit bands, reliability flags, competition columns, and the product disclaimer. - Shareable links that preserve the brand, weights, filters, and shortlist. ## How the method works ### 1. Describe every neighborhood Each of the 83,008 census tracts in the continental United States is described using 21 features. These cover income, home values and rents, education, population density, age, renter share, vehicle access, commute mode, household composition, race and ethnicity, poverty, and surrounding retail context. Retail context includes the density of points of interest, grocery stores, restaurants, cafes, and shopping. ### 2. Learn the brand's archetypes Store locations are mapped to the neighborhoods they occupy, using one observation per occupied neighborhood so colocated facilities do not double-count. A chain may thrive in several genuinely different neighborhood types, such as dense urban cores and affluent car-oriented suburbs. LookAlike therefore clusters occupied neighborhoods into as many as three store types and creates a separate archetype for each. The product names these store types and shows the share of locations in each. ### 3. Score by nearest archetype Every neighborhood is scored by its weighted distance to the nearest store-type archetype. Using the nearest archetype avoids the distortion caused by averaging genuinely different store types into a profile that represents none of them well. Results are reported as both a similarity percentile against the whole country and an absolute Strong, Moderate, or Weak fit band measured against the brand's own store neighborhoods. The absolute band prevents a high national percentile from hiding a mediocre absolute match. ### 4. Stress-test the answer For any neighborhood, the robustness check re-scores the result using equal weighting and with each factor theme removed. This separates stable ranks from ranks that depend on one theme. Neighborhoods with small populations or large income margins of error receive a low-reliability badge derived from the source data's published error margins. Factor weights remain visible and adjustable, and they are saved with every shared scenario. ## What it measures and what it refuses to claim LookAlike measures: - Observable neighborhood resemblance using open demographic and map data. - Where a brand's existing footprint pattern repeats, including markets the brand has not entered. - Open white space relative to that footprint pattern. - Competitor proximity based on tracked chains. LookAlike does not measure: - Revenue, sales, foot traffic, or whether a specific store would succeed. - Rents, zoning, co-tenancy, visibility, or other site-specific factors. Those require the user's own diligence. - Complete store-location coverage for every chain. Coverage varies by chain, and the application states a caveat where relevant. Demographic resemblance explains only a modest share of store performance. That is why LookAlike is a discovery and triage tool rather than a prediction system. It ranks where to look; the user's team still decides. ## Who it is for ### Retail and restaurant expansion teams Turn "where next?" into a ranked, explained list of open markets before paying for site-selection studies on the wrong metros. ### Franchise development Show candidates the neighborhoods that look like proven territories, with the supporting evidence one click away. ### Commercial real estate brokerage and tenant representation Enter a pitch with a defensible market screen based on the tenant's own footprint, plus a clean report that the brokerage can present under its own name. ## Frequently asked questions ### What is retail site selection? Retail site selection is the process of deciding where to open new store locations. In practice it moves through three stages: screening markets (which metro areas deserve attention at all), evaluating trade areas and neighborhoods (which parts of those markets fit the concept), and site-specific diligence on individual properties (rents, zoning, co-tenancy, visibility, access). LookAlike works at the first two stages: it screens 383 US metro markets and ranks all 83,008 US neighborhoods by resemblance to the places where a brand already operates. Site-specific diligence stays with your team. ### How do I evaluate a neighborhood for a new store location? Start from evidence you already have: the neighborhoods around your existing successful stores describe the conditions your concept thrives in. Then compare candidate neighborhoods against that profile across three groups of characteristics: demographics (income, home values, rents, education, age, household composition), density and urban form (population density, renter share, commute mode, vehicle access), and retail context (the density of nearby grocery stores, restaurants, cafés, and shopping). LookAlike automates exactly this comparison, describing every US neighborhood with 21 open-data features and showing a per-feature explanation for every candidate it ranks. ### What data does LookAlike use? Open data. Neighborhood characteristics come from US Census American Community Survey data at the census-tract level, plus map-derived retail context such as the density of points of interest, grocery stores, restaurants, cafés, and shopping. Store locations for built-in brands come from public sources, and coverage varies by chain; the app states a coverage caveat wherever it applies. You can also upload your own store list, which stays in your private workspace. ### Does LookAlike predict sales or revenue? No, and it says so inside the app and on every export. LookAlike measures neighborhood resemblance: how much a candidate neighborhood looks like the neighborhoods where a brand already operates. Demographic resemblance explains only a modest share of store performance, which is why LookAlike positions itself as discovery and triage rather than prediction. Tools that promise revenue forecasts from analog matching are claiming a precision the data does not support. ### What is an analog store model? An analog model (also called a lookalike model) is a long-standing retail technique: profile the locations where a concept already succeeds, then search for places that resemble them. LookAlike is a transparent, national-scale implementation of that idea. It clusters a brand's occupied neighborhoods into up to three store types, builds an archetype for each, and scores every US neighborhood by weighted distance to the nearest archetype, with every weight visible and adjustable. See the method for the full walkthrough: https://lookalikeiq.com/method/ ### How is LookAlike different from a demographics report? A demographics report describes one place you already picked. LookAlike ranks every neighborhood in the country against your footprint, so you start from a national shortlist instead of checking places one at a time. And every score explains itself: a similarity percentile, an absolute fit band, a per-feature fingerprint against your brand's archetype, a robustness check, and a reliability flag on noisy estimates. ### Can I use my own store list? Yes. Upload a CSV of your locations and rank the whole country against your real footprint. You can add a performance column to weight the archetype toward your best stores, or flag closed and underperforming locations so scoring becomes resemblance-to-winners minus resemblance-to-failures. Uploads become versioned datasets in your private workspace and can be reloaded without re-uploading. ### Can I compare two brands? Yes. Compare puts two brands on one diverging map and shows which brand each neighborhood leans toward, with three rankings: leans you, resembles both, leans them. It works between two built-in brands or between a built-in brand and your own uploaded footprint, which makes it useful for sizing contested ground against a competitor. ### Who is LookAlike for? Teams that put pins in maps for a living: retail and restaurant expansion teams deciding which market comes next, franchise development teams showing candidates territories that look like proven ones, and commercial real estate brokers and tenant reps who need a defensible market screen for a client pitch. ### Is my uploaded data private? Yes. Uploaded store lists live in your private workspace, are not shared with other customers, and are used only to score your own brand. You can export or delete your data at any time. ### How do I get access? LookAlike is in a closed beta with a small number of retail and commercial real estate teams. Request access with your work email and we will reach out as seats open. Beta teams get direct input on the roadmap: https://lookalikeiq.com/request-access/ ## Guides ### What is site selection? A practical explanation of the three-stage site-selection funnel: screen markets, evaluate trade areas and neighborhoods, then conduct diligence on individual properties. It identifies the questions, evidence, roles, and common failure modes at each stage, and explains where software supports triage while human judgment remains essential. https://lookalikeiq.com/guides/what-is-site-selection/ ### Evaluating neighborhood characteristics A guide to defining trade areas, using census tracts as a consistent neighborhood proxy, and comparing candidates across demographics, density and urban form, and retail context. It explains why existing stores should define the starting profile and why survey margins of error and reliability flags must remain visible. https://lookalikeiq.com/guides/evaluating-neighborhood-characteristics/ ### The analog store method A transparent account of analog modeling: describe neighborhoods with 21 features, learn store-type archetypes, score candidates by nearest-archetype distance, and stress-test the result. It states the method's proper role in discovery and triage and explains why neighborhood resemblance cannot predict sales, revenue, foot traffic, or store success. https://lookalikeiq.com/guides/analog-store-method/ ## Access LookAlike is onboarding a small number of retail and commercial real estate teams during a closed beta. Request access: https://lookalikeiq.com/request-access/ Contact: hello@lookalikeiq.com