All examples

Case study · Insurance brokerage

Finding insurance brokers
with a real shopfront.

AI agents screened more than 10,000 business premises to find brokers with a visible shopfront. Every firm on the final list had supporting evidence and was checked by hand.

1 min 10 · Original French animation with English subtitles

The
challenge.

An insurance group is growing by acquiring independent brokerages. Before approaching a firm, its team needs to know who it is, where it is and whether it is still trading.

Business directories mix head offices, registered addresses, insurer agents and firms that have closed. Sorting them by hand takes weeks. By the time the list is ready, parts of it are already out of date.

The task was to find every broker in the area with an actual shopfront, and keep evidence for each one.

Give AI
the right tools.

Ask an AI assistant this question on its own and it will produce a list that is partly made up. It needs tools to look up the records, inspect the premises and keep track of what it finds. Here is how those tools fit together.

How
it works.

  1. Start with the official register

    The agents query the INSEE business register for active establishments registered as insurance agents or brokers. That gives them more than 10,000 premises in the area, with an address for each.

  2. Look at the premises

    For each address, they retrieve the most recent street image and its date, then check the business’s Google listing.

  3. Classify what is visible

    AI classifies each image: a shopfront with signage, an office, a house or an unclear view. It looks for insurer logos and signs identifying a broker. A second pass checks every shortlisted result with a specific yes-or-no question.

  4. Keep the evidence

    A firm only makes the list if its shopfront and sign are visible in a dated image and its Google listing is active. I then review every firm on that list myself.

  5. Put it on one page

    A private page brings the results together: a map, a list of firms and, for each one, the evidence and links to its street image and Google listing.

The
results.

10,000+
business premises screened
10,446
building fronts inspected by AI
139
firms verified, each reviewed by hand

There are limits. Street images can be several years old, and the camera often points at the business next door. Against an independent reference list, the system found just over half the agencies. Firms without a visible shopfront are kept for further checking, rather than ruled out.

What you
could build.

You can apply the same approach to lists of prospects, suppliers, shops or construction sites. Start with an official source, let agents check each entry and keep the evidence. Then review the results that matter.

This is the kind of tool we can build together in our sessions, using a task from your business. You do the building, with me there to guide you.

Let’s talk about your business

What’s taking up too much time?

Tell me what keeps slowing you down. We’ll look for a practical starting point, whether that’s less admin, better follow-up with prospects or a smoother service for your customers.

Book a call to get started