Skip to main content

The Death of the Big 4: AI-Enabled Services Are Opening a Whole New Market

When I started in venture a decade ago, software categories typically had 2 or 3 contenders gunning to win. As software’s gotten easier to build (and flooded by ample venture dollars), competition has skyrocketed in almost every market. The deluge of me-too products has made it hard for buyers and founders to tell the difference. Don’t get me wrong, there's major opportunity remaining in software. But it might come in a different package. 

Services businesses are a promising next frontier. With massive market sizes- the Big 4 alone brought in $200B in revenue last year- and a relative dearth of dynamic tech talent, the ground is fertile. Historically, common wisdom has held “services businesses aren’t venture backable.” Compared to software businesses, services performed by humans tend to be low margin, hard to scale exponentially, lumpy in revenue, etc. etc. There are many cautionary tales. But change is afoot… hastened by AI, cracking open new, huge opportunities for founders brave enough to go against the tired wisdom. 

We even have a beacon out there showing the way: Palantir. What started as a services business is now amongst the most highly-valued (on a multiples basis) venture-backed company in the industry. There’s much to be learned from its march to higher margins and recurring revenue. And with the rapid advancement of AI, I’d argue we’re on the verge of a Cambrian explosion of Palantirs.

Service businesses that leverage both AI and humans to deliver holistic solutions to clients are poised to outgun and outpace the services behemoths that’ve dominated for the last 50 years. Legacy providers are ripe for disruption: their business models rely on human labor and hourly billing which can be turned on its head by AI-enabled vendors. Further, their “product” (humans) is very difficult to integrate AI into. While software incumbents are having a relatively easy time adapting to the GenAI wave, services incumbents will struggle. Founders, this is a boomtown moment to seize.

For a long time now, founders pursuing software exclusively have been strip mining for the same, highly competitive gold.

There’s a bounty to be had by the few companies building disruptive AI-enabled services. Think this:

We want to help by sharing the potential we’re seeing out there. We’ll delve into which types of services are ripest for disruption, and go into detail on how startups can modify their approaches to product, GTM, and more to build the strongest companies possible in the space.

These are the services most ready for change

Service firms are hired for two reasons: 

  • Do a job that the client doesn’t have the bandwidth or expertise to do.
  • Offer third-party expertise in decisions (cynically, to cover the client’s a**.)

That execution-oriented first bucket tends to include IT implementations (like cloud migration projects), financial audits, and outsourced customer support – things clients want a services firm to do for them. 

The second more advisory bucket is home to M&A banking services, strategy consulting, and wealth management – things clients want a services firm to help them with.

We think AI-enabled services businesses can thrive in that first bucket. Startups have the chance to combine humans and software to handle those “do it for me” tasks that can be easily automated and have objective outcomes. 

The more advisory-focused a project is, the less a client will trust a bot. The better opportunity for startups in this category is to play more of an enabler role- selling AI-enabled software to help existing service providers. There’s some golden opportunities here too – think of the software that could be sold to law firms, banks, Deloitte itself (!) to help them do more with less. 

So we’ve got the “Help you do it vs. do it for you” dichotomy. The other big one is “repeatable vs. bespoke.” The more repeatable a service, the more productizable. The more productizable, the better the startup opportunity. 

We used these two spectrums to build a framework that makes it clear where we see the best opportunities to capture services market share. 

Disruptors reading this today, look to that top right quadrant. Those are the low-hanging, productizable fruit.

One bonus consideration: It’s worth studying the services that have high failure rates and been difficult to perform in the past. Can AI do much better? Then it’s worth pursuing. 

For example, we just invested in a company called Mechanical Orchard that uses AI to move data trapped inside clunky mainframes to the cloud. This was not just hard – it was often impossible – to do before. The code was just too old, broken, and dense with no way to map it. 50% of migration projects driven by services firms failed. Mechanical Orchard saw this and put AI to work, teasing this code apart, and making it safe and secure to move. 

The way we see it, services startups going after execution-oriented, repetitious tasks are going to out-deliver the competition to scale quickly and efficiently– and that’s very venture-backable.

How to build one of these companies

These hybrid companies – part AI-enabled software, part people – don’t have a lot of precedent. There are a lot of open questions: Can AI-enabled solutions perform well enough? How to build a trusted brand when clients are already skeptical? Do you hire the same types of people or not? 

We’ve aggregated what we’ve learned alongside the early leaders in the space, like Mechanical Orchard and Harvey. But we’re all definitely in learning mode, so we’d love to hear your perspectives. 

Here are the emerging best practices we’ve seen:


  • Fewer jobs-to-be-done. Focus on just 1-2 jobs-to-be done that are shared by many clients. The more jobs, the harder to productize. You can always expand.
  • Scalable access. Find and tap into high-quality talent channels. Mechanical Orchard did this with the Pivotal Labs network (CEO Rob Mee used to run that company). Where’s your leg up?
  • Piggyback software. Focus on those opportunities where you can sell software alongside services. AI will widen margins by improving service delivery, but true software sales will anchor repeatable revenue.


  • Believable brand. Services firms are often hired based on relationships and reputation. Startup founders need to invest in building the same. 
  • Product marketing FTW. It’s on these marketers to figure out what customers really want and need, and translate that into messaging that slam dunks the value of AI for the client. Speed? Cost savings? Whatever it is, it has to be punched hard. 
  • Befriend your frenemies. Be strategic in how you navigate the existing ecosystem of services firms – they’ve been around for a while and some of them may be great channel partners if you can help them solve a challenge for their clients that they aren’t equipped to. 
  • Outcome-based pricing. Charging for the value your work unlocks instead of for the inputs (like labor) allows you to capture more value beyond what you put in (like what the AI does…) This might not be feasible before you demonstrate value, but you can move to it over time.

Org Design

  • Hybrid humans. The new worker this trend creates will have skills on both the client side and the product side. When devs aren’t on client work at Mechanical Orchard, they’re building internal software. The versatility helps tighten the product feedback loop. 
  • Honed metrics. Your metric tracking needs to be bulletproof from the beginning, both for GAAP accounting purposes and for operational decision making. 
  • Aligned incentives. Each team’s incentives need to be aligned with evolving staffing and pricing models. You want to reward folks for building more versatile skill sets and selling services and product that produce great outcomes.

We’ll add to this list and host events that dig deeper into what’s changing across organizations (and people’s skill sets) as AI takes root in new ways. If you’ve observed any of this, or other best practices, we’d love to hear from you. 

Blurring the line between something and someone

While the software industry has come to consume the world, it’s important to remember that software is just a means to an end. As Clay Christensen reminds us, buyers really just want to make a hire that gets the job done well — whether that’s hiring someone or something.

While software and services firms have evolved as separate entities, it would be easier for a buyer to have a single vendor to hold responsible. And as AI blurs the line between that “something” and “someone,” the era of the full-service startup may well be upon us.

The implications of this are both simple and predictable, and potentially wild and far flung. On the very likely to happen side of the spectrum, it stands to reason that if companies can do a whole job-to-be-done (instead of capping out because they only focus on software OR services), they’ll be able to extend their TAM further than ever and capture more value. We’re already seeing this underway.

On the wilder end, just think – what if a full job-to-be-done firm enabled companies to outsource everything except their core competency. For example, a firm could presumably handle all of your go-to-market from marketing to sales, customer experience to success, in which case you could focus solely on building and refining your core product. If this was real, how much faster would technology move? It’s fun to think about.

Bringing it back to the present, we’re excited to track the burgeoning crop of companies exploring new combinations of software and services with the help of AI — as well as the ripple effects created by this big shift. If you’re building something that fits these descriptions, we’d love to hear from you. Send us a note anytime at or