Nobody Knew How Many Property Management Companies Exist. So We Counted Them One at a Time.
Peter Lohmann asked a question we've all had. It struck our curiosity too, so we pulled every property management business listing Google knows about in 98 US markets, looked at their websites and reviews, and checked 1,006 of them by hand. Here is what we found.
Updated September 2026. This replaces our July pilot, which covered 24 markets and reported ~33,000. The number came down. The whole reason why is below.
A Number Everybody Uses and Nobody Has Checked
In August 2026, Peter Lohmann put out an estimate that there are around 31,000 residential property management companies in the country. He got there another way: start with a published total of professionally managed rental doors, then divide by roughly how many doors an average company manages.
That is a sensible approach and it is the best one available if you are working from public data. The catch is that the answer depends on the average you pick, and nobody has ever gone and checked that average against the actual companies.
We were curious about the total too. We have worked with hundreds of property management companies, and our gut said the number of real, working operators was smaller, maybe 10,000 to 14,000. So instead of debating the average, we tried the boring version: go find the companies and count them.
The answer landed at about 24,800, between the two. Lower than the door-division estimate, higher than our own instinct. Both of us were somewhat wrong, which is the most useful outcome available.
Goodjuju is a marketing agency for property management companies. A bigger industry is better for us, so read our headline with that in mind. It is also why we are publishing the whole method below, including the corrections that made our number smaller and the mistakes we made along the way. Peter and I are friends and talked while this was going on. This is meant to sit next to his work, not to argue with it.
About 24,800 Companies, With a Real Range Around It
Two methods with nothing in common, landing about 6,000 apart. Peter's figure falls just outside the top of our range, which is a real disagreement rather than agreement, and worth more than if we had matched.
Counting upward from real businesses gives a smaller answer than dividing total doors by an average company. That is the finding. It suggests either that the average company manages more doors than assumed, or that the published door total includes property that is not managed by the kind of company either of us is trying to count, or some of both.
We cannot settle which from our side alone. But the gap is consistent, it is larger than our margin of error, and it is the thing worth arguing about.
Our Pilot Said 33,000. This Says 24,800. Here Is Why.
In July we published a pilot covering 24 markets and reported about 33,000. That page is gone and this one replaces it. Three things changed, and two of them were our mistakes.
1. We searched four times more of the country
The pilot covered 24 markets holding 7.1% of America's renter households. This covers 98 markets and 59.9%. More importantly, the pilot leaned on very few small markets, and one of them was Nantucket. A resort island where property managers look after second homes for absentee owners has far more managers per year-round renter than a normal town. With only three small markets in the sample, Nantucket was effectively setting the rate for every small market in America. We added 34 more, chosen at random before we looked at any of their results, and that distortion went away.
2. We had been counting the software's guesses, not the truth
The software sorts every business into yes, no, or not sure. It is wrong a fair amount in both directions. The pilot number leaned too heavily on its raw "yes" count. The correct approach is to hand-check a sample from every group, including the ones the software rejected, and use those measured error rates to correct the total. When we did that properly, the number came down.
3. We fixed a bug that was inflating the total by thousands
This one is embarrassing and we would rather say it than have someone find it. To correct for the software's errors we compare its verdicts against hand-checked samples. Some companies were re-checked after the sample was drawn and moved from "yes" to "no". Because those companies had been selected for checking at a much higher rate than ordinary "no" companies, counting them as ordinary "no" companies made the "no" group look far more error-prone than it is, which inflated the final number by roughly 7,000. We had built a safeguard against exactly this months earlier, and then wrote a piece of analysis that walked around it. There is now a test that fails if anyone does it again.
We are showing the arithmetic on our own revision because a study that only ever moves in the direction its author would prefer is not worth much. Ours moved down, twice.
Fitting the Definition and Being Findable Are Two Different Things
When you count companies one at a time, you can do something an average cannot: look at each one and ask how much business is actually there. Of the roughly 25,000 real companies, here is what an owner searching for a manager would actually be able to find.
About one in five of these companies has no customer reviews at all, and about one in ten has no website.
Separately, we opened a sample of the websites to see whether they still load. About 21% were dead. So the number of companies actively competing for new owner business is meaningfully smaller than the number that exist.
Most of What Google Files Under Property Management Is Something Else
We pulled 159,304 Google listings from the property management category and merged them into 112,636 separate businesses. Only a minority turned out to be third-party residential property managers. About 65% were rejected outright: apartment complexes renting their own units, commercial brokers, HOA-only managers, realtors, and a long tail of businesses in the category for no discernible reason.
The starkest group we called ghost listings. A Google Business Profile in the property management category with no website, no reviews, and no trace anywhere on the web when we searched the exact company name. Rather than guess what they were, we checked 135 of them by hand, one at a time, with the Google profile and the search results open.
What they actually were:
This is the part you can only see if you go company by company.
If you divide total doors by an average, every company in your answer is a real one managing an average number of units. There is no way for the math to tell you that a chunk of the category is an airport parking lot. You have to open them up and look.
What We Treated as a Property Management Company
Every call in this study runs against one set of rules, used the same way by the software and by the person checking by hand. A company counts if it manages residential rental property for other people who own it, meaning single-family homes and small multifamily, as a real staffed service.
A company with a real, staffed third-party residential division counts here even if they also manage mostly big apartment buildings. Peter's definition looks at a company's core business, which would leave some of those out. Neither is more correct, and honestly I like his better. We were each after a slightly different number, and it is worth knowing which one you are reading.
The Process, Start to Finish
What we are counting is companies, not doors. Everything below is about turning a pile of map listings into a number of businesses you can stand behind.
Pick markets that look like the country
98 markets out of all 2,225 US metro and micro areas, sorted into six size bands so the mix runs from New York and Chicago down to single rural counties. Every band has at least twelve markets in it, which matters: in our pilot the smallest bands had only three apiece, and one resort island skewed the whole national figure.
Search Google Maps until it runs dry
Six search terms run across a grid of towns and cities inside those markets, going page by page at every single location until the results stop turning up anything new. Not a fixed number of pages: we keep paging until a page returns nothing we have not already seen. That gave us 159,304 listings, including the ones with no website attached.
Merge the duplicates
One business often has several pins. We combined them by franchise office, then website, then phone number, then name and state. Roll-up brands collapse into one company; franchise offices stay separate because they are independently owned. 159,304 listings became 112,636 companies.
Read whatever evidence exists
We read each company's website, starting with their owner-facing pages. When there was no website, or the site would not load, we pulled their Google reviews instead. Whether the reviews come from owners or from tenants turns out to be a strong clue about who a company actually works for.
Sort every company against the rules
All 112,636 companies were sorted by AI against the rules above, and every decision was saved with its reasoning and the quotes it came from, so any single call can be looked up and checked later.
Check the software's work by hand
1,006 companies were reviewed by hand, drawn on purpose from every group including the ones the software had thrown out, and weighted so the over-sampling does not bias the result. The review screens showed the evidence but never showed what the software had decided, so the human answers could not be nudged by it.
Fix the errors, then scale up
We applied the error rate measured in each group, added a correction for companies Google never showed us at all, and scaled from our 98 markets to the whole country using renter households, band by band rather than all at once.
Roughly 1 in 11 businesses Google files under property management is a third-party residential property manager. The software's own count was about twice that, which is why the hand-checking step decides the answer.
Checking Whether Google Showed Us Everybody
The fair objection to a study built on Google listings is simple: what about the companies Google does not show you? So we measured it. Take a few lists of property management companies that were built with no help from Google at all, then see how many of those companies our search actually turned up.
| Outside list | Companies | We found | Why it is a fair test |
|---|---|---|---|
| Franchise directories | 52 | 90–92% | Built from franchise filings |
| NARPM members | 28 | 89–100% | An association membership list |
| AppFolio / Buildium customers | 32 | 94% | Built from who bought software |
| All together | 112 | 88–95% | Corrected for in the headline |
A franchise directory and a software customer list have nothing to do with each other except that neither one came from Google. They agree, which is the useful bit.
Matching is harder than it sounds, and getting it wrong would have changed the answer a lot. Of the NARPM members we confirmed, only about two thirds matched on the website listed on their profile. Five had no website on their profile at all and we found them by phone or name. Four more had moved to a completely different website since joining: one member's profile pointed at his personal realtor site while his management company traded under a different name entirely. If we had simply matched website to website, we would have reported finding 57% of the list instead of 89%, and concluded our search had a coverage problem it does not have.
One thing nearly caught us out here. The software list first looked like we had only found 64% of it, which would have been a real problem. Then we looked at the misses: a real estate private equity firm, a company doing veterinary real estate, a commercial retail landlord. Our search was right to skip those. But you cannot throw out the off-topic companies from the misses only, because that quietly cleans up one side of the math and makes your result look better than it is. So every company on the list went back through the same rules, without us knowing which ones we had found. Once that was done, we had 30 out of 32.
All three of those lists are lists of companies that joined or bought something: a franchise, a membership, a software subscription. Lists like that lean toward established, professional companies, and those are also the easiest ones for Google to know about. So 88–95% is the best case for our coverage, not the worst. And no extra list can fix that, because no list of buyers contains the companies that never bought anything.
How Often Was the Software Wrong?
Software that has never been checked against a person is an opinion, not a measurement. So 1,006 companies were reviewed by hand, and we deliberately pulled from the piles the software had rejected too. A study that only double-checks its own yes answers can never find out what it threw in the bin. In this study the rejected pile is where most of the argument lives, because it is by far the biggest.
Dick's Sanitation Service
The review tool, showing a real company from the pile our software had already rejected. Dick's Sanitation is a garbage collection company that Google files alongside property managers, and the software was right to throw it out. We go through the rejected pile anyway, because the only way to find out whether it threw away anything real is to look at what it threw away.
| What the software said | Companies | Checked by hand | Actually real | Real companies |
|---|---|---|---|---|
| Yes, has a website | 12,928 | 195 | 53.3% | 6,895 |
| Yes, no website | 6,365 | 180 | 12.2% | 778 |
| Not sure, has a website | 14,851 | 250 | 11.2% | 1,663 |
| Not sure, no website | 4,805 | 23 | 4.3% | 209 |
| No, has a website | 64,765 | 291 | 0.7% | 445 |
| No, no website | 7,988 | 40 | 0% | 0 |
We applied each group's measured error rate to that whole group. The software said yes to 19,293 companies in these markets; hand-checking says about 10,000 are real. It is too generous on its yes answers and slightly too harsh on its no answers, and the two mistakes do not cancel.
Look at No, has a website: 64,765 companies, of which we hand-checked 291 and found 2 real ones. That 0.7% is the single most important number in this study. The group is so large that being wrong by one percentage point moves the national figure by about 2,000 companies. It is also the most boring work in the project, which is why we put the most hand-checking into it and why our earlier draft got it wrong.
The corrections that moved the number
The first hand-check was done in a very wide spreadsheet that meant scrolling sideways to read review text. It produced enough errors that we threw it out and built a proper review tool instead: one company per screen, with its details, its reviews, a link to its website and a link to its Google profile. Both times we compared the two passes, the errors ran the same way: the spreadsheet made us reject real companies, never accept junk.
The larger mistake was the one described further up this page, where a piece of our own analysis walked around a safeguard we had built for exactly that problem and inflated the answer by roughly 7,000 companies. We are telling you because this is exactly the kind of thing that quietly ruins a study, and because the only reason we caught it was going back through our own work looking for it rather than waiting to be asked.
A Lot of These Companies Are Not Really Trading Anymore
Once you have the companies in front of you, you can ask something an average never can: is this business still running? Two separate signals say a fair number are not.
Google reviews, across every company the software counted
Nearly a third have never had a single Google review. Another third have fewer than ten. Quiet reviews on their own do not prove much, and plenty of good managers go a year or two without a new one, so we checked a second signal as well: whether each company's website still loads. Of 1,422 counted companies we opened, 292 (21%) came back dead or parked.
Where we were able to pull the actual review dates, about 60% had a review in 2026 and roughly one in nine had nothing since 2022. That part is a sample rather than a full count, because pulling review dates costs a separate lookup per company and we only did it where we needed the evidence to make a call.
Put the two signals together and you get the split further up the page. A live site plus some recent activity means a company that is working. Nothing on either usually means the listing has outlasted the business.
Decisions That Changed the Number
A count like this is mostly definitions. These are the calls that moved the number the most. Each one was decided once and then applied to everything. Flip any of them and the total changes, which is why they are here in the open instead of buried in a footnote.
Working Down From Doors, or Up From Companies
| Doors divided by an average | Counting companies (this study) | |
|---|---|---|
| Starts with | A published total of managed doors | Individual business listings |
| Rests on | How many doors an average company manages | What you decide counts as a company |
| Best at | Giving a fast, repeatable national figure anchored to real published housing data | Producing an actual list you can ask follow-up questions about |
| Cannot see | Junk listings, or how big each company is | Companies with no public presence at all |
| Answer | 31,000 (range 26,000–36,000) | 24,762 (range 20,700–29,600) |
Neither one replaces the other. Peter's approach is quick, repeatable, and tied to a real published door count from federal housing data. Ours is slow and expensive but ends with a list of companies, which is the only way to ask things like "how many of these are still trading?"
The useful result is that they disagree, by about 6,000 companies, with the two ranges only barely touching. If we had matched Peter's number it would mostly tell you that two methods can be tuned to the same answer. A gap this size says something is going on, and there are only a few candidates: the average doors per company may be higher than assumed, the published door total may include property managed by companies neither of us is trying to count, or our search may be missing more small operators than our coverage test can detect. Our money is on some of each.
We cannot settle it from our side alone, and we are not trying to. Peter's method has a real advantage ours does not: it is anchored to federal housing data rather than to what Google happens to know.
What This Study Cannot Tell You
- The range is wide and will stay wide. 20,700 to 29,600. Most of that comes from one number: what share of the 64,765 companies our software rejected are secretly real. We measured it at 0.7% on 291 hand checks. Because that group is enormous, being off by a single percentage point moves the national figure by about 2,000 companies, and tightening it meaningfully would take thousands more hand checks, not hundreds.
- A third of the answer comes from small markets. The bands below mid-size metros supply about 31% of the total, from twelve markets each. That is far better than the three apiece our pilot used, but it is still where this study is thinnest, and it is the first place we would push if we were reading someone else's version of this.
- We can only count companies with some public trace. A manager with no website, no reviews and no Google listing is invisible to us. It is a real boundary, and the coverage test below cannot see past it either.
- Coverage was tested against lists of joiners. 88 to 95% is the best case. Every list we could check against, franchises, an association, software customers, is a list of companies that joined or bought something, and those are also the easiest for Google to know about. No list of buyers contains the companies that never bought anything.
- Resort markets do not behave like normal ones. Places like Nantucket have far more managers per year-round renter, because their customers are second-home owners who do not show up in the renter-household count we scale by. We now have enough small markets that this no longer distorts the national figure, but it is a real wrinkle in the method.
- We are not a neutral party. We sell marketing to property management companies. A bigger industry suits us. The whole method is here, including two of our own errors that made the number smaller, so you do not have to take our word for any of it.
Three Things Worth Knowing
Counting upward gives a smaller industry than dividing downward
About 24,800 companies, against 31,000 from the door-division method. If you are sizing a market, hiring against it, or building something for property managers, it is worth knowing that the two respectable approaches differ by roughly 20%, and that nobody has yet reconciled them.
Most of your competitors are not really competing
Nearly a third of the companies our software counted have never had a single Google review, and about a fifth of the websites we opened were dead or parked. The number of companies actively going after new owner business is meaningfully smaller than the number of companies that exist.
Being findable is most of the game
We could not verify a company that left no trace online, and an owner looking for a manager cannot either. Every one of those 135 ghost listings was invisible to us for exactly the same reasons it would be invisible to a prospect.
The full data and method are available
Every decision, every classification, the coverage tests and the review tools. If you want to check our work, or you are researching this industry, get in touch.
Ask for the methodologyStart With Your Free Custom Video Analysis and Game Plan
Check out the responses from others after their video analysis:
Hello Landon,
Thank you so much for your video email. Your video was informative, and I really appreciate you in taking the time to send us a video (it is always nice to see an actual face with someone we do business with as well!)
In your video, you were mentioning that the quote that you would work for us is…
Hey Landon,
We would like to move forward with The Takeover option. Would you be able to add 6 city pages instead of 5 for the package at the same price? We are currently in those 6 cities now and would like to keep that footprint.
Can you let me know what next steps would be to get this started?
Watched the video, very interesting and agreed with everything.
Let's get going on the website for sure (the middle package you recommended). I'm interested the SEO as well but need to see how to fit it into my budget.
Let me know what next steps are.
Hi Landon,
Thank you for taking the time to put together the video! I watched it and I appreciate your transparency as well as your respect for our time. You raised some very salient points, and I will go over them with the owner of the company. It is ultimately up to him whether…