How private equity teams screen a city for a roll-up of independent businesses
A roll-up needs many owner-run businesses close together, few chains and room to improve. How deal teams read a city's mix before the first call to an owner.
Part of the guide Workflow automation
A private equity team screens a city for a roll-up by asking whether it holds enough owner-run businesses of one kind, close enough together to share managers and back office, with few chains already buying them up and enough spread in quality that better running would show. xMap's places database answers the first three directly: how many there are, how many belong to a chain, and how their ratings spread. The fourth, whether owners will sell, is the deal team's job, but the data tells them which owners to meet first.
This is for deal teams, operating partners and search funds who build platforms in dental, veterinary, salons, physiotherapy, accounting or trades. It shows how to read a city's mix, works through dentists in three cities with real counts, describes a weekly AI employee that keeps the target list current, and lists the readings that mislead.
What makes a city worth a platform
- Depth. Enough independents of the kind that a platform of ten or twenty sites is a small share of the market, not most of it.
- Fragmentation. A low chain share. Where groups already own many sites, prices for the rest are higher and the best owners have had offers.
- Density. Targets within a short drive of each other, so one regional manager and one supply route can serve several.
- Room to improve. A wide spread of ratings. A market where everyone is rated 4.9 gives a buyer little to fix; one with a cluster of well-placed, middling businesses is where operating changes show.
- Signs of change. Few openings can mean a settled market with ageing owners; many openings mean new competition for whatever you buy.
Reading dentists in three cities
xMap's places database lists 830 dentists within 10 km of the centre of Toronto, 768 in the same circle around Sydney and 638 around London (November 2026). The chain and rating figures below are worked out from the dentists nearest each centre, not from every one listed.
| City | Dentists listed | Nearest counted | Independent | Chains | Rated 4.5 or more |
|---|---|---|---|---|---|
| Toronto | 830 | 460 | 447 | 13 | 313 of 366 rated |
| Sydney | 768 | 465 | 447 | 18 | 297 of 337 rated |
| London | 638 | 462 | 429 | 33 | 261 of 329 rated |
Within 10 km of each city centre, xMap's places database, November 2026. Independents, chains and ratings are of the nearest dentists counted in each city.
All three are fragmented: of the 460 dentists nearest the centre of Toronto, only 13 belong to a chain, and Sydney's 465 nearest include 18. London's nearest 462 hold 33 chain practices, more than twice Toronto's, which suggests groups have been buying there longer. The rating spread is narrow in all three cities, with most rated practices at 4.5 or more, so in dentistry the improvement case rests more on scheduling, purchasing and hygiene staffing than on fixing poor reviews.
Openings differ too. Among Toronto's nearest dentists, 42 of the 457 with an opening date opened in the twelve months before November 2026, against 5 of 462 in Sydney and 7 of 460 in London. A team weighing Toronto should expect new practices competing for the same patients and the same associates.
Building the target list
Once a city passes the first read, the team needs a list it can work: every independent of the kind, where it is, how it is rated, how many people rated it, its website and phone, and which other targets sit near it.
- Add an Area of interest with the city, or draw the metro area you would run from one office.
- Add Nearby places with the type (Dentist, Veterinarian, Beauty salon and so on), Independents only, and Most places set high enough to cover the city.
- Add a second Nearby places on Each incoming place, with counts, within 3 km, same type. Each target gains how many others sit within a short drive: its cluster.
- Add an xMap LLM block that scores each target for fit with your thesis, for example 'well placed, average rating with many reviews, neighbours within 3 km', and explains its score in one line.
- Add Sort and limit to put the highest fit first and Show on a map so the clusters are visible.
- Save the list to a sheet with an app block, or to your own database with Save to my database, so the team's notes sit beside it.
Put a Schedule on it every week once a deal is live, and add What changed before the email. It reports targets that closed, changed name or saw their rating move, which is often the first sign that an owner is retiring or that a rival group has bought in.
A made-up example of the cluster score
The following is made up to show how the cluster count reads. Say a team screening veterinary clinics in Denver finds 60 independents. Twelve of them sit within 3 km of at least four others, in three tight groups along the main roads. Those twelve become the first meetings: buying one gives a reason to meet its neighbours, and a regional manager could visit all of a group in a morning. The other 48 stay on the list for later, sorted by fit.
Readings that mislead
- Treating listed as independent. A practice owned by a group but trading under its old name looks independent in any places data. Check ownership before an offer.
- Comparing a total with a sample. A city page's chain share is of the places nearest the centre. Do not divide it by the city's total count.
- Ignoring places with no rating. Many small practices have no rating at all. They are not bad; they are unrated, and some of the best-run owners never ask for reviews.
- One search for a whole region. A single search from a region's middle finds only what is near it. Name each city or draw each zone.
Questions
Can it tell me which owners are ready to sell?
No. It shows who exists, where, how they are rated and what changed. Signs such as a closed second site or a rating drop can prompt a call, but readiness comes from talking to owners.
Which kinds of business work for this?
Any kind the places database lists by type: dentists, vets, physiotherapy, beauty salons, opticians, accountants, plumbers and many more. Pick the type from the list in Nearby places.
Can I compare several cities at once?
Yes. Put each city in the Area of interest and the list carries an area field, so a chart or a sort by city shows depth and chain share side by side.
How do I keep targets we already met off the list?
Keep a table of them in My data and add Match with my table, set to keep only the places not found.