AI / Tech-Enabled Roll Ups are a Dumb Idea

How did this even become a thing

Once in a while, a new VC investing theme pops up that makes you scratch your head. For the past year, one popular theme / debate has been “tech-enabled rollups”, which is a flavor of the Private Equity rollup strategy but with a twist.

Basically, the idea is that you buy a bunch of “boring” businesses in non-tech industries with healthy cashflow, and inject a bunch of “tech / AI tools / agents” to streamline the business to increase top-line and bottom-line.

Before explaining why this idea is dumb, let’s level set on why some VCs love this idea.

The whole bet is that AI will deliver a meaningful uplift for the business, and that tech bros know how to run a plumbing business / accounting firm / law firm / car wash / building cleaning / lawn mowing business / doctor’s offices, etc, better than someone who’s been doing it for decades.

The hope, of course, is that an AI-”augmented” cleaning business (to pick an example) will be so much better than a crusty one running on phones and Excel. Eventually, the AI version dominates the local cleaning industry, which allows the VC to also sell its AI CRM, voice agents, vertical agents, etc, to other cleaning businesses to earn product revenue as well.

Then voila, the VC builds a Toast 2.0 or Square 2.0 for cleaning industry, and build an unicorn just like that.

Or, worst case, the VC can just keep making money with their cleaning business, giving them downside protection even if the AI play doesn’t work out. Heck, maybe the VC can occasionally help out with cleaning to improve bottom line. This creates some embedded optionality, which of course VCs love. This quote from a TheInformation article captures the sentiment:

“You have to buy the accounting firm,” which can be more expensive than giving a founder a seed investment check of $1 million or $2 million, said ChenLi Wang, a general partner at WndrCo. “But if it doesn’t work, at the end of the day you still have an accounting firm,” whereas software businesses that don’t achieve a high enough scale usually have to shut down entirely.

ChenLi said his firm’s roll-up investment strategy was inspired by Berkshire Hathaway and Constellation Software, which operate portfolios of businesses they’ve bought.

But perhaps the darker side to this “Tech-Enabled Rollup” play is a silent admission by many vertical-focused VCs that making and selling AI agents or AI SaaS is a crappy business in general, at least compared to the glory days of SaaS from 2000 - 2021.

The main hardship of selling AI agents is that these products have little differentiation, little buyer urgency, and too much competition for the popular use cases (e.g. customer support). So this Tech-Enabled Rollup strategy is basically trying to be “cute” and buy the customer wholesale, so it can dogfood its own AI products to make “AI revenue”.

But now onto why this idea is dumb.

Let’s all take a biiiiig step back and just use some common sense.

First, there’s already a massive ecosystem of tech vendors already catering to boring, crusty businesses with “tech”. Setting up websites, managing their IVRs, helping with Wordpress, spinning up chatbots, and operating CRMs. AFAIK, many of these local tech vendors are starting to pitch AI agents onto their existing customer base. So it’s not like these tech enabled rollups have no competition whatsoever.