Services-Tech
The half that gets rebuilt
I spent four articles describing something I did not have a name for.
The name arrived sideways, in a text exchange with one of my partners. He was not talking about positioning. He was thinking about how a prospective investor in our fund at Bambu Capital would tell us apart from another manager he knows, one who also buys services businesses and has done well at it.
His conclusion took two lines.
That firm invests in services companies that will not be transformed by technology. We invest in services companies that will be.
Read that again, because it is not a marketing distinction. It is a description of two genuinely different assets that happen to share an industry code.
The last four pieces were the argument. This one is the label, and why the label does real work.
The phrase everybody uses says nothing
Ask any mid-market private equity firm what it invests in and a good number will say tech-enabled services. I have said it myself, in rooms where I should have known better.
The problem is that it describes a condition rather than a category. Tech-enabled is a state a company is currently in. It tells you the business uses software. It does not tell you whether technology is about to rebuild the way that business makes money, or whether it will simply keep making the same money slightly more efficiently.
Those are not variations of the same investment. They are opposite bets on the same balance sheet.
One is a productivity story. The company does what it already does, with better tools, at a somewhat better margin. Value comes from operating discipline, from buying well, from consolidating a fragmented market. This is a real and often excellent way to make money. It is most of what private equity has done in services for thirty years.
The other is a transformation story. The way the company produces its output, prices it, and staffs it all change inside a few years. Value comes from being early to a business model that does not exist yet at scale. The upside is larger. So is the chance of being wrong.
A phrase that covers both is not a thesis. It is a shrug.
What the other categories got right
Fintech did not mean a bank that bought software. It meant the moment technology rebuilt how money moves, and it named a set of companies that could not have existed before.
Proptech did the same for how buildings get bought, financed and managed. Insurtech for how risk gets priced and underwritten. Healthtech for how care gets delivered.
Each of those names did three things at once. It told you which industry was being rebuilt. It told you that the rebuilding was the point rather than a feature. And it gave capital, talent and founders a word to organise around, which is most of how a category becomes real.
Services-Tech is the same event arriving at the businesses that sell expertise.
Consulting. Systems integration. Agencies. Advisory. Managed services. Clinical services. Anywhere the product is a person’s judgment and the invoice is denominated in hours.
That is a large piece of the economy, and until now the only word for what is happening to it has been a description of the weather.
The definition, and the line it draws
Services-Tech is the set of services businesses whose economics get rebuilt by this technology rather than merely improved by it.
The complement matters just as much, and I want to be careful to say it without condescension, because it is where a great deal of money has been made and will continue to be. Call it services non-tech: businesses that sell expertise where the work, the pricing and the staffing model survive the arrival of AI substantially intact.
Plenty of excellent businesses sit on that side. Trades. Field services. Anything where the constraint is a licensed human physically present. If technology is not going to reprice the core of what you sell, being disciplined about buying and running those businesses is a perfectly good way to compound capital.
The mistake is not choosing one side. The mistake is holding a portfolio that straddles both without knowing which one you bought.
How to tell which side a business is on
The four previous pieces were, without my realising it at the time, a diligence list. Four questions decide it.
Does the invoice survive? Part two was about the firm that had AI running through its whole delivery chain and no pricing model to capture any of it. If a business bills by the hour for work that now takes a fraction of the hours, every efficiency it introduces is a tax on its own revenue. That is a Services-Tech problem whether the owner has noticed or not.
Does the pyramid survive? Part three was about the base of the labour pyramid. If the business grows by hiring juniors and billing them at a multiple of their cost, and the rote work that justified those juniors is now automated, the growth engine has to be rebuilt rather than tuned.
Does the catalogue survive? Part four was about the second wave, the one almost nobody prices. A firm whose offerings exist because a platform is hard to implement is exposed to that platform’s roadmap. A catalogue is a list of things that used to be difficult.
Can it say where the judgment sits? Part one was about governance. A firm that has thought carefully about what it lets the machine decide, and what it keeps for people, is a firm that has actually engaged with the change rather than bought a licence and put out a press release.
Answer all four honestly and you know which business you are looking at. In my experience the owner usually knows before the buyer does, and does not always volunteer it.
Why an investor should care more than a marketer
Here is the part that made me take my partner’s text seriously rather than treating it as a slogan.
The two categories carry different risks, and they carry them at different times.
A services non-tech business carries execution risk. It might be badly run, it might be bought at the wrong price, the end market might soften. Those risks are legible, and the industry has thirty years of practice pricing them.
A Services-Tech business carries all of that plus timing risk, and timing risk is the one that is genuinely hard. As I wrote in part four, a firm in the middle of a real transformation looks almost identical to a firm failing at one. Both show compressed margin and investment landing ahead of return. Wait for the evidence and it is priced. Move before it and you might be buying the failure.
That difference should show up in how the deal is structured, how the rollover is set, how long the hold is, and what the underwriting assumes. Two funds can buy businesses with the same industry code, the same revenue and the same margin, and be running completely different risk models.
An investor deciding between two managers deserves to know which one they are getting. Not because one is better, but because they are not substitutes.
Some investors want exposure to the services economy as it is. Some want exposure to the part of it that is about to be rebuilt. Those are different products, and a phrase that covers both serves neither.
The uncomfortable part
I should be honest about the weakness of the name, since I am the one proposing it.
Fintech, proptech and healthtech each name an industry. Services is not an industry, it is most of the economy, so Services-Tech is broader and vaguer than its siblings. That is a fair objection and I do not have a clean answer to it. The closest precedents are adtech and martech, which name a function rather than a vertical and became perfectly useful anyway.
The other honest caveat is that a category name is a claim, not a fact. Fintech means something because thousands of people used it the same way for a decade. One person writing it in five articles is a phrase. Whether it becomes a category depends on whether it turns out to be useful to people who owe me nothing.
I think it will be, for one reason. The distinction it draws is one that owners, buyers and investors are all currently making with clumsy language, and everybody in those conversations can feel the gap.
Where this leaves the series
Four articles argued that something is happening to the businesses that sell expertise by the hour. The economics break first. The labour model breaks second. The catalogue breaks third, and almost nobody is pricing that one. Underneath all of it sits a question about what a firm still keeps for human judgment.
This piece says the plainest thing I can about what to call the result.
Not every services business gets rebuilt. The ones that do are a category, they are investable, and they carry a different risk than the ones that do not. At Bambu Capital we are standing on the rebuilt side deliberately, and underwriting the timing risk that comes with it. Being explicit about which side of that line you are on is more useful than another year of everybody saying tech-enabled and hoping the other person means the same thing.
Some services businesses get improved.
Some get rebuilt.
Only one of those is a bet on timing.
A six part series on what AI is doing to professional services.
One. AI Decides How. Humans Decide What, Why, and If. Governing AI inside a single firm, and where the human veto has to sit.
Two. Nobody Is Buying Hours Anymore. What happens to an industry that priced and sold the one thing the machine turned out to be best at.
Three. Guess Who Becomes AI Native. Whether this wipes out a generation of young professionals, and what becomes of the pyramid.
Four. The Offerings Transform Too. The second wave, where the platform makes your catalogue unnecessary, and why productising is the wrong answer.
Five. Services-Tech. You are here. Naming the category, and the line between the services businesses that get rebuilt and the ones that do not.
Six. Nobody Is Ripping Anything Out. The AI layer over the systems you already own, the migration nobody has named, and why it is the first good news in the series.