The venture capital convinced themselves that they have found the next big investment advantage: use AI to wipe the software type margins in the high intensity of traditional workforce. The strategy consists in acquiring mature professional service companies, implementing AI to automate tasks, then using improved cash flows to remove more companies.
The General Catalyst Catalyst (GC), which has devoted $ 1.5 billion, devoted $ 1.5 billion to its last fundraising to what it calls a “creation” strategy which focuses on the incubation of native software companies AI in specific verticals, then using companies as acquisition vehicles to buy established companies – and their customers – in the same sectors. GC has placed bets in seven industries, legal management services, with plans to extend to up to 20 sectors.
“Global services are a turnover of 16 dollars per year worldwide,” said Marc Bhargava, who directs GC’s related efforts in a recent interview with Techcrunch. “In comparison, software represents only 1 dollars billion in the world,” he noted, adding that the attraction of software investment has always been its higher margins. “As you get software on a scale, there are very few marginal costs and there is a lot of marginal income.”
If you can also automate service activities, he said – by attacking 30% to 50% of these companies with AI, and even automation of 70% of these basic tasks in the case of call centers – mathematics are starting to seem irresistible.
The match plan seems to work. Take Titan MSP, one of General Catalyst portfolio companies. The investment company provided $ 74 million on two tranches to help the company develop AI tools for managed service providers, then it acquired RFA, a well -known IT services company. Thanks to pilot programs, said Bhargava, Titan has shown that he could automate 38% of typical MSP tasks. The company now plans to use its improved margins to acquire additional MSPs in a conventional withdrawal strategy.
Likewise, the firm incubated Eudia, which focuses on internal legal services rather than law firms. EUDIA has signed Fortune 100 customers, including Chevron, Southwest Airlines and Stripe, offering legal services at fixed costs powered by AI rather than traditional hourly billing. The company recently acquired Johnson Hanna, a legal service alternative, to extend its scope.
The general catalyst seeks to double – at least – the margin of Ebitda of the companies it acquires, explained Bhargava.
Techcrunch event
San Francisco
|
October 27-29, 2025
The power firm is not alone in its thought. The venture capital company Mayfield has cut $ 100 million specifically for AI “teammate” investments, including GROVE, a computer consulting startup which acquired a security consulting company of $ 5 million, then increased it to 15 million dollars in revenues within six months while reaching 80%gross margin, according to its founders.
“If 80% of the work will be carried out by AI, it can have a gross margin of 80% to 90%,” Navin Chaddha, general manager of Mayfield, Techcrunch this summer. “You could have mixed margins from 60% to 70% and produce a net income from 20% to 30%.”
The Solo Elad Gil investor has been pursuing a similar strategy for three years, supporting companies that acquire mature businesses and transforms them with AI. “If you have the assets, you can [transform it] Much faster than if you just sell software as a supplier, “said Gil in an interview with Techcrunch this spring.
But the signs of early alert suggest that all the metamorphosis of the service-industries can be more complicated than the VCs do not provide. A recent study by researchers from Stanford Social Media Lab and the Betterup laboratories who interviewed 1,150 full -time employees in all industries revealed that 40% of these employees had to exhaust more work because of what researchers call “workslop” – works generated by AI which seem polished but lack substance, creating more work (and helmets) for colleagues.
The trend is wreaking havoc on organizations. The employees involved in the survey say that they spend an average of almost two hours dealing with each Workslop body, including to decipher it, then decide to send it back or not, and often just to repair it themselves.
Based on the estimates of the participants of the time spent, as well as their self -depressed wages, the perpetrators of the survey believe that Workslop has an invisible tax of $ 186 per month per person. “For an organization of 10,000 workers, taking into account the estimated prevalence of workslop … This reports more than $ 9 million a year in loss of productivity,” they write in a new article in the Harvard Business Review.
Bhargava challenges the notion that AI is superchy, asserting rather than all these implementation failures in fact validate the approach of General Catalyst. “I think that shows the opportunity, that is to say that it is not easy to apply AI technology to these companies,” he said. “If all the fortune 100 and all these people could simply bring in a consulting company, slam on an AI, obtain a contract with Openai and transform their business, then obviously our thesis [would be] A little less robust. But the reality is that it is really difficult to transform a business with AI. »»
He underlined the technical sophistication required in AI as the most critical missing puzzle piece. “There are a lot of different technologies. It’s good in different things,” he said. “You really need these AI engineers applied from places like ripple and ramp and figma and scale, who have worked with the different models, understand their nuances, understand which are good for what, understand how to wrap it in software.” This complexity is exactly the reason why General Catalyst’s strategy of combining AI specialists with industry experts to create companies from zero is logical, he argued.
However, it is undeniable that Workspolop threatens to undermine the main economy of the strategy. Even if a Holding company is created as a starting point, if the companies acquired reduce the staff, because the AI efficient thesis suggests that they should, they will have fewer people available to catch and correct the errors generated by AI. If companies maintain current endowment levels to manage the additional work created by the problematic production of AI, the huge margin gains on which VC count could never be made.
It is easy to argue that these scenarios should perhaps slow down the scaling plans which are at the heart of the strategy of withdrawn from the VC and which potentially undermine the figures which make these transactions which are attractive to them. But let’s face it; It will take more than one or two study to slow down most investors in Silicon Valley.
In fact, as they generally acquire companies with existing cash flows, General Catalyst says that its “creation strategy” companies are already profitable.
“As long as AI technology continues to improve, and we see this massive investment and improvement in the models, I think there will be more and more industries to help incubate companies,” said Bhargava.
