FlowsBook a callSign inCreate free accountSign up
AI employees and automation

An AI employee that tells you when something opens in your street

Set an AI employee to watch a few hundred metres around your shop and message you only when a new business of a kind you care about appears. What to set and what to expect.

Part of the guide AI employees

By the xMap team · · 5 min read

An AI employee can watch your street for you: you give it your shop's position, a short distance and the kinds of business you care about, and it checks on a timetable you choose. When a new cafe, salon or pharmacy appears within that distance it sends you the name, what kind of place it is, how far away it is and its details. On the weeks nothing changes, it stays quiet.

This is for owners of one or a few shops, restaurants and clinics who want to know about a new neighbour before customers mention it, and for anyone who looks after a street: a landlord, a business improvement district, a local newspaper. It explains the two ways to set it up, which settings matter, a made-up example in Melbourne, the mistakes to avoid and what arrives in the inbox.

Two ways to watch a street

1. Run only when something changes. The When something changes trigger starts the flow only when a check finds a change. You set Watch for to a new place of a kind opening nearby, pick up to four kinds from xMap's places database, choose up to five places to watch around and the distance, and how often to check: every 15 minutes, hour, 6 hours or day. The first check only looks; from then on, each new place of those kinds within the distance becomes an item and the rest of the flow runs.

2. A weekly sweep. A Schedule every Monday, an Area of interest with your shop as a place and a radius, Nearby places for the kinds you care about, and Only new so each place is reported once. This suits a wider area or more than four kinds, and gives you a single message a week.

Both report places that are new to the data. A place that has been open for months but was only just added can appear as new, and a place that opened yesterday may take a while to be listed. Where a place carries an opening date you can see it in the message; many places do not have one.

The settings that matter

  • Distance. For a street, 200 to 500 m. Wider, and you hear about places your customers will never compare with you.
  • Kinds of place. Your direct rivals first. A bakery might watch bakeries, cafes and convenience stores; a nail salon, beauty salons and nail salons.
  • How often. Once a day is plenty for most shops. Every 15 minutes is for teams that act within the hour, such as a sales rep chasing new openings.
  • Where it reports. An email, a WhatsApp message, or a note in the Team inbox where you can ask it follow-up questions.

A made-up example in Melbourne

Invented for illustration. The owner of an independent coffee shop in a Melbourne laneway sets the trigger to watch for new cafes and bakeries within 400 m of the shop and of the tram stop at the end of the lane, checking once a day.

For three weeks nothing arrives. In the fourth, one evening, a message: a new cafe has appeared 180 m away, part of a chain, with a phone number and website but no rating yet. xMap AI's line suggests checking its opening hours and whether it sells pastries, because the owner's morning trade relies on them. The owner walks past the next day, sees it opens an hour later than her shop, and starts promoting her early opening on the board outside. She adds the newcomer to her own list of rivals so a weekly sweep can show how its rating develops.

Mistakes to avoid

  • Watching too far. A 2 km radius in a city centre sends news about places that do not matter to you, and you stop reading.
  • Too many kinds. Pick the kinds that take your customers. A cafe does not need to hear about every new dentist.
  • Reading new to the data as just opened. A place may have been trading for a while before it was listed. Check before you react.
  • Expecting a message every week. Quiet weeks are normal on most streets. The flow is working when it is silent.
  • Forgetting to look at the first message carefully. It sets your expectations. If it reports places that are not rivals, narrow the kinds.

What the AI employee does, block by block

  1. When something changes: Watch for a new place of a kind opening nearby, the kinds you care about, your shop and one or two nearby points, 400 m, check every day.
  2. xMap LLM: tell it what your shop sells and ask for one line on whether the newcomer competes and what to check.
  3. Email to you, or a WhatsApp Business step, or both.

When a new place of those kinds appears within the distance, you get one short message: its name, the kind of place, whether it is part of a chain, its distance from your shop, its address, phone and website, its rating when it has one, and the AI's line on what to check. The flow also keeps each newcomer as a result, so the Team inbox and the map show every new neighbour over time. On weeks with no change you receive nothing.

Questions

How far around my shop should it watch?

For a street, 200 to 500 m. Choose the distance a customer would walk to the alternative.

How many kinds of business can it watch?

The When something changes trigger takes up to four kinds and up to five places to watch around. For more, use a weekly sweep with Nearby places and Only new.

Will it tell me the day a business opens?

It tells you when a place first appears in the data, which can be before or after the real opening. Places with an opening date show it.

What if nothing opens for months?

Then you receive nothing. The trigger only runs the flow when a check finds a change.

Can it also tell me when a rival closes or its rating drops?

Yes. The same trigger can watch for a place closing or a rating falling below a line. Use a second AI employee for each so the messages stay clear.

Templates that do this

Read next

More on ai employees and automation · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.