How lenders check the area around a small business loan applicant

A lender checks the area around a small business by looking at the applicant's address the way a customer would: how many businesses of the same kind are within a short walk, how many opened or closed lately, how well they are rated compared with the applicant, and how busy the area is. Those few facts do not decide a loan, but they explain a business plan's assumptions or contradict them, and they take minutes rather than a site visit.

This article is for small business lending teams at banks, credit unions, fintech lenders and microfinance institutions, and for the credit analysts who write the file note. It covers which facts to pull, how to read them fairly, a worked example, what not to do with them, and an AI employee that writes the area note for every new application.

The area facts that matter for a loan

• Direct competitors. Businesses of the applicant's type within 300 to 500 m, from Nearby places counting around the applicant's address, with the nearest one's distance.
• Chains next door. Whether the nearby rivals belong to chains. A new independent cafe next to two chain coffee shops has a harder first year than one among independents.
• Ratings around it. The average rating of the rivals. If the applicant already trades, its own rating and number of reviews against theirs.
• Churn on the street. Openings in the last 12 months and places that closed. A street where many have opened and closed is riskier than a stable one.
• How busy the area is. Footfall per day and its trend from Mobility Footfall, in the cities it covers.

How to read the facts fairly

Read them against the plan, not against a fixed rule. A plan that assumes no competition on a street with eight rivals needs a better answer about what is different. A plan that assumes heavy passing trade in an area with low footfall needs a closer look at where customers will come from. When the facts agree with the plan, the note says so and the file moves on.

Use the same distance for every applicant of a kind, so notes are comparable across the portfolio. And keep in mind that a dense street can be a good sign: restaurants and cafes often cluster because the customers are there.

What a dense market looks like

City figures show how crowded some markets are before you even look at a street. xMap's places database lists 512 bakeries within 10 km of the centre of Toronto; of the 369 nearest the centre, 37 belong to a chain and 332 are independent, led by COBS Bread with 8. In London the same circle lists 3,111 cafes; of the 342 nearest the centre, 5 opened in the twelve months before November 2026.

512Bakeries listed, Toronto

332Independent, of the 369 nearest

3,111Cafes listed, London

5Opened in the last year, of the 342 nearest

A new independent bakery in central Toronto joins a market of mostly independents; one asking for a loan in central London's cafe market enters a crowded field where few newcomers appeared in the past year. Neither fact is a yes or a no; both shape the questions.

A worked example

Here is a made-up example. An applicant wants a loan to open a second bakery in a Toronto neighbourhood. The analyst enters the address, and the area note comes back: within 400 m there are four bakeries, one a chain branch, rated 4.3 on average; one opened in the past year and none closed; the area's footfall per day is above the middle of the applicant's first site's area, and its trend is flat. The applicant's first bakery is rated 4.7 from several hundred reviews.

The analyst's note reads: competition is real but not new, the applicant out-rates its future neighbours, and the street is stable. The questions for the call are about rent and staffing, not about whether customers exist. The decision still rests on the accounts.

Mistakes to avoid

• Declining because a street is busy with rivals. Clusters often mean demand. Read the count with the ratings and the churn.
• Using the facts as a hidden score. Write them into the file note where the applicant's officer and reviewer can see and question them.
• Applying one rule to every country. Footfall coverage, place density and chain shares vary. Compare an applicant with its own city.
• Forgetting the existing book. The same facts, refreshed monthly for current borrowers, warn early when a big rival opens next door.

Doing it with an AI employee

Applications usually arrive as rows in a sheet or a table. A flow started by New rows takes each new application, then: Nearby places (counting rivals of the applicant's type around its address, chains, ratings, opened in the last 12 months), Mobility Footfall (footfall per day and the trend), an xMap LLM block that writes a five-line area note from those figures only, and Email to the analyst on the file. What lands in the inbox, minutes after the row appears, is the note, the figures under it and a small map.

For the existing book, a monthly Schedule with My data (current borrowers) and What changed reports borrowers whose neighbourhood moved. Starting points: Check new merchants from my sheet, Growing small companies for my business loans team and Weekly small business pulse.