Shortlists & reports

Saving neighborhoods to a shortlist

Click the star icon on any neighborhood — either in the map popup, the candidates list, or in the detail panel — to add it to your shortlist. The same neighborhood can be saved under multiple brands, keyed by the combination of brand and tract; switching brands does not erase neighborhoods you have already saved.

When you are signed in, your shortlist is saved on the server under the retail client you are working in, so it follows your account rather than one browser and colleagues with access to the same client see the same starred neighborhoods. The first time you sign in, any shortlist already sitting in that browser is imported once into the account. In the unauthenticated local development mode, the shortlist falls back to browser local storage (lookalike-shortlist-v1, with older bae-shortlist-v1 data migrated automatically).

Either way, the current brand’s shortlist is also encoded in the URL (the pin parameter), so a shared link — or the Copy link button — carries your starred neighborhoods to whoever opens it.

What each saved item records

When you star a neighborhood, the tool captures:

The neighborhood’s tract GEOID and name
The brand you were viewing when you saved it
The similarity percentile at that moment (so if you revisit, you can see what it was when you saved it, even if you later change factor weights)
The factor weights in effect when you saved it (your decision context)
A text note field where you can record why you are interested in it, who to call, next steps, or any other context

You can edit the note for any saved neighborhood at any time.

Exporting a PDF report

Once you have saved at least one neighborhood, the “Export report (PDF)” button becomes active in the shortlist panel.

Clicking it opens a new browser tab with a print-ready, single-page-per-neighborhood PDF. The report includes:

Neighborhood header: rank, name, similarity percentile vs. the saved brand
Feature comparison table: 8 key demographics for the neighborhood vs. the brand’s average (median household income, education, density, age, rental rate, vehicles, home value, rent)
Top resemblance drivers: the 3 features most like the brand’s archetype (if the match is modest, the heading becomes “Closest features”)
Your note (if you wrote one)
Tract GEOID: the Census identifier
A flag if you used custom factor weighting when you saved it

Footer and disclaimer: The report restates that it measures observable neighborhood resemblance only — not revenue, foot traffic, or store success. It names the data sources (ACS 2023 5-year, TIGER/Line tract boundaries, OpenStreetMap stores).

You can print or save as PDF from your browser’s print dialog. The report is styled to avoid orphaned rows and sections across page breaks.

Exporting a CSV

The “CSV” button exports your entire shortlist as a comma-separated-values file suitable for further analysis in a spreadsheet or data tool.

The file includes:

One row per saved neighborhood
Columns: geoid, label, brand, saved_similarity_percentile, fit_band, low_acs_reliability, note, and all 21 features used in the tool (demographics and retail POI density)
Feature values are rounded to 2 decimal places
A disclaimer row at the top: “LookAlike shortlist — discovery/triage only; NOT a sales or revenue prediction.”

The ranked candidate list has its own CSV export (from the lookalikes list, not the shortlist panel) that additionally carries nearest_store_km, nearest_competitor_km, and competitors_within_6km for competition screening.

Formula injection safety

The CSV uses defensive escaping: any non-numeric cell that begins with =, +, -, @, or a tab character gets a leading single quote prepended, so spreadsheet applications treat it as text rather than a formula. Plain numbers — including negative values such as continental-US longitudes — are left untouched so they stay computable. Cells containing commas or quotes are quoted and internal quotes are doubled.

How saved items relate to custom uploads

If you upload your own store list, you can also save neighborhoods to a shortlist in that custom mode. The saved items will have brand: "custom" and the name you gave your upload. You can later export reports and CSVs that mix built-in brands and your own uploads, but remember that custom-mode similarity is computed differently — it reflects your own store performance data if you provided it.

Resetting your shortlist

Removing a neighborhood is immediate and there is no undo. When you are signed in, the removal is saved to the account, so it applies for everyone working in that retail client. In the unauthenticated local development mode the shortlist lives only in that browser, and clearing browser storage erases it.