# A method you can see all the way through.

LookAlike's site-selection method: 21 open-data features per neighborhood, store-type archetypes from your real footprint, transparent distance-based scoring, robustness checks, and reliability flags on every score.

How a match is computed

LookAlike scores resemblance with a transparent measure over open data. Every step is visible in the product, and every number traces to its source.

1.  ## Describe every neighborhood
    
    Each of the 83,008 neighborhoods in the continental United States is described by 21 features covering income, home values and rents, education, density, age, renter share, vehicle access, commute mode, household composition, and the surrounding retail context.
    
2.  ## Learn the brand's archetypes
    
    A brand's stores are mapped to the neighborhoods they occupy, one observation each. Because a chain often thrives in several distinct kinds of place, the occupied neighborhoods are clustered into up to three store-types, each with its own named archetype and share of locations.
    
3.  ## Score by nearest archetype
    
    Every neighborhood is scored by its weighted distance to the nearest store-type, so a single averaged archetype cannot rate real store neighborhoods as mediocre. Scores are reported as a national similarity percentile and an absolute fit band measured against the brand's own stores.
    
4.  ## Stress-test the answer
    
    The robustness check re-scores a neighborhood under equal weighting and with each theme removed, separating stable ranks from ones that hinge on a single theme. Noisy estimates carry a low-reliability badge, and your weights stay visible and travel with every scenario you share.
    

Neighborhood fingerprint · vs Trader Joe's

Brand archetypeThis neighborhood

99.9th percentile

Strong fit

Tract 84.03, Sacramento County, CA

Resembles Trader Joe's suburban / lower-density stores

Why it matches

-   Households with children26.8%
    
    Close to the typical 26.7%
    
-   Population density1,864/km²
    
    Close to the typical 1,877/km²
    
-   Average household size2.39
    
    Close to the typical 2.39
    

Where it differs

-   Median age36.0
    
    Lower than the typical 40.8
    
-   Walk / bike / transit1.3%
    
    Lower than the typical 7.1%
    
-   Restaurants19.5/km²
    
    Higher than the typical 12.1/km²
    

## Questions about the method

### What does a resemblance score tell me?

How closely a neighborhood matches the places where your brand already works, measured across 21 features. It is a strong, explainable signal of where your footprint repeats — the ranked starting point your team investigates first.

### Can I exclude race and ethnicity from the model?

Yes. One click removes those features from the score, and the app explains the trade-off so the choice is deliberate and recorded.

### How do I know which scores to trust?

Noisy small-area estimates carry a low-reliability badge, and a robustness check shows whether a rank survives equal weighting or hinges on a single theme.

## Put the method to work.

Pick a plan, upload your store list, and rank every US neighborhood against the places you already win.

[Get started](https://lookalikeiq.com/pricing/)[See the method](https://lookalikeiq.com/method/)

Canonical URL: https://lookalikeiq.com/method/