Turning Customers into Investments
Vertical SaaS buyouts – how software actually eats the world
A provocative question has been circulating among software investors in recent months:
Is there alpha left in software as an asset class?
The question is more often than not raised by venture capitalists (VCs), who have spent the past two decades pouring significant resources into horizontal SaaS platforms, covering nearly every conceivable application. The rationale behind these investments is straightforward: only markets with substantial scale can potentially deliver the power law returns that VCs chase.
However, this perspective seems more like a convenient narrative than an accurate assessment of the current landscape. It neglects the numerous verticals—let’s call them offline industries—where software adoption has lagged despite obvious advantages. These sectors typically involve operating models centered around real assets and services where technology is not a core competency. This suggests a vein of untapped opportunities for software investors – but they’re often in smaller, underserved markets that have resisted technology. I would argue that there is alpha to be found in applying software to many verticals, but distribution strategies need to be reimagined.
The challenge and opportunity lie in shifting focus from horizontal to vertical SaaS (vSaaS). Historically, most VCs have shied away from vSaaS because these markets appeared too small to support the kind of outsized returns VCs require. However, standouts like Toast, which debuted with a $20 billion market cap, show that vertical markets can indeed be venture scale bets by rolling out new playbooks for TAM expansion.
In 2022, Bessemer Venture Partners introduced the concept of 'layer caking' – the idea that TAM can be expanded by adding complementary services or products. For example, Toast started with POS and subsequently added reservations, payroll, and financing but stayed focused on the restaurant vertical. Over the past few months, I’ve noticed an emerging playbook that could redefine how investors and entrepreneurs approach vSaaS.
Note: despite the contentious preamble, I still think there is alpha in horizontal software and in markets that have been early adopters of technology. AI (and future tech waves) will continue to open the door to build better products. However, we’re left with two options: continue to replace legacy solutions in markets that have previously adopted technology or try and rethink distribution altogether.
Vertical SaaS buyouts
A few months back, Yoni Rechtman wrote about a software adoption thesis on Twitter.
The basic gist was this:
There are many industries where value should accrue to software, but they've traditionally been resistant to tech
New entrants come along and say "if we build a better product, they'll use it this time" – this happens every tech wave
However, if there's a culture / adoption problem, a better product doesn't solve this
This would mean these companies could (likely) capture considerable value by adopting the right software
Thesis is vSaaS buyouts: build the software yourself then begin acquiring operating companies/customers
You capture the full value from the P/L transformation and combat adoption risk
Use a mix of cash flow from operating assets, debt facilities, and additional equity to finance further acquisitions
The example he used to illustrate it was Metropolis, a computer vision parking lot management application. This thpiqued my curiosity.
Will Manidis dropped a similar thought around a month later:
Now the idea had my attention. It has been rattling around in the back of my mind since.
Applying software to industry is a major lever for accelerating GDP growth, yet the pace of this growth is directly tied to the rate of technology adoption. This issue is particularly pronounced in AI. Despite ongoing discussions over the past year about the transformative potential of AI/ML, the reality is that the impact of these technologies on our daily lives depends on how quickly they are adopted. I’m increasingly convinced that vSaaS buyouts are how software actually eats the world.
Case studies
I’ve been talking to investors and entrepreneurs across the market to find examples of this playbook being rolled out. It’s hard to identify a large list of players actively rolling up a vertical like this–holding your cards close to your chest is strategically advantageous. There are a handful of players that publicly fit the bill:
Metropolis: Metropolis is the leading example of this playbook in action. They developed a computer vision application for managing parking lots and has expanded by acquiring parking lot assets.
Eli Dukes at Verticalized has written a great case study on their strategy
Electric Sheep Robotics: manufacturing robotic lawnmowers and is acquiring landscaping services to deploy its technology after the acquisition.
Splash: rolling out a reverse playbook, Splash acquires operational pool service companies and enhances them through a blend of existing and newly developed software to optimize operations.
Sinovation Disrupt Fund: They acquire operational businesses and have a team of ML developers to parachute in and build custom ML solutions tailored to these companies.
Carbon Health: EHR provider, rolls up primary health clinics
LedgerUp: built an AI bookkeeper
Tangentially, this playbook lends well to Sarah Tavel’s ‘Sell work, not software’ thesis
Hippo: InsurTech company – uses sensors + AI for home monitoring – acquired Spinnaker Insurance
Hadrian: Chris Power is developing software to automate machine shops within the U.S. defense supply chain and is integrating vertically with these factories.
Patrick O'Shaughnessy discusses this further in his podcast with Chris Power and Josh Wolfe
Note: while Hadrian’s approach deviates from the typical playbook since they are building their own factories, it began with a roll-up strategy.
Market selection
I think there are a handful of defining characteristics of verticals where it makes sense to run this playbook. This is an expanding (and malleable) list on the market dynamics to look for:
Adoption Risk: Markets where there has traditionally been low software adoption.
Undervalued Potential: Industries where software should logically add value, but this potential has not yet been realized.
Market Fragmentation: Industries that are fragmented, lacking dominant players.
Operational Simplicity: Acquisition targets have relatively straightforward operating models.
Scalable Testing: A normal distribution of business sizes allows the strategy to be tested on smaller scales before escalating to larger, more complex acquisitions.
Data aggregation: specific to AI-first software (our focus), I’m looking for markets where owning operating assets provides a data aggregation advantage.
Note: these are easier characteristics to describe than to find!
Executing this strategy at a fund-level scale may be complex, but there is a unique niche for micro-SMBs. Traditionally, these smaller markets have been difficult to penetrate with software due to long sales cycles, small total addressable markets (TAMs), and low average contract values (ACVs), which hurt capital efficiency. However, by owning the underlying assets, there's a clearer path to capturing value and enhancing software integration in these industries.
Two ways to play
There are two main ways you can roll out this playbook – software first or operator first.
Software First
In the ‘Software First’ approach, the focus is on developing a robust software solution before acquiring or integrating with operating businesses. This path allows for refining technology to meet specific industry needs without the initial complications of managing a non-tech business.
Metropolis is a case study for the software first model. They built a computer vision application for parking management before acquiring parking assets. This approach allowed them to perfect their technology in a controlled setting, ensuring that the software could genuinely enhance operational efficiency before dealing with the complexities of physical asset management.
On one hand, this allows the entrepreneur to focus on one thing at a time–concentrating on development without the distraction of running an operational business. On the other hand, if the founder does not have industry experience, they run the risk of developing a solution that does not fully address the practical challenges of the industry. In Mark Leonard’s words, they might lack the ‘earned secret’ that comes with industry experience.
Operator First
The ‘Operator First’ strategy involves acquiring or establishing a foothold in the operational side of the industry before developing or customizing software solutions. This approach allows the entrepreneur to live the problem before designing a solution.
Splash is an example of the operator first approach by acquiring pool service companies and then integrating new and existing software to manage them more efficiently. This strategy allows them to directly experience the operational challenges and tailor their technology solutions in real time.
My gut feeling is that this doesn’t allow for the same leverage as the software first approach. Simultaneous demands of managing operations and developing software can strain resources and focus. However, it does combat the risk of hubris. It’s a bold assumption to think that you can run a business more efficiently just because you have the right tech stack. Being an operator first gives you a taste of humble pie.
Financing strategy
Luke Sophinos brought up an interesting question recently – This playbook strikes a middle ground between the power law risk profile of traditional venture investing (software component) and the smoothed risk profile of private equity (operating asset portfolio) – so who funds it?
A few interesting conversations have shown me the possibilities:
I think Appa at Toro has the right take. In the long run, there may be a middle ground asset class that emerges. But for now, it’s different capital providers. The early part of the journey requires a capital partner willing to finance software for smaller TAMs since this playbook lends to building software for vertical niches. You may need to raise additional capital for the software, but once it’s shipped, you need to build a more traditional capital stack to invest in operating assets (private credit, mezz, etc). This likely bifurcates your corporate structure into two vehicles, lending to an OpCo / PropCo structure. You raise equity against the software co (OpCo) and then raise subsequent equity and debt facilities against the operating assets you acquire in PropCo. Two different risk profiles, two different asset classes – two different investor types are needed… for now.
OpCo / PropCo was popularized by the commercial real estate market. More recently, some forward-thinking PropTech investors have been playing with this idea as well. Alpaca VC has done some great thought leadership here – see ‘The Nuts and Bolts of PropCo / OpCo’ by Daniel Fetner.
The short-term winner: Studios
This playbook has a venture return opportunity without the absolute downside. We haven't seen an asset class like that in a long time and it won’t be long before new capital players emerge to fund it. In the short run, I think that venture studios have a unique opportunity to capitalize on this. Studios have a more flexible capital allocation mandate than most VCs and can steer incubated software companies toward this playbook from Day 1.
The risks
The largest risk that I see with this playbook is the hubris associated with thinking you can run a business better because you have the right tech stack. These businesses are far more complex than that and the magic is often in the quality of services provided. There are two things that players considering this strategy must be wary of:
The right people – you need the right operators to run the business once you’re in. This also includes a killer integration team that can roll out the right tech / playbook and to the handle volume of acquisitions as you scale
Degrading pride – owner / operator businesses often have a great quality of service due to the pride associated with ownership. As you scale, companies need to look at strategies in retaining that quality of service. Otherwise you’re playing a zero sum game of financial engineering. You capture long term value by owning the business and delighting the customer – don’t forget this.
Multiple hats: this playbook is much more difficult than traditional SaaS. You’re effectively wearing two hats as a founder – product builder and capital allocator. You almost need a dedicated founder or team member with M&A experience to run the allocation business.
Let’s build
This is an open call to entrepreneurs building companies using the software-first approach to this playbook. If you’re currently building or have an idea where this makes sense, I want to talk to you. We are an AI-focused venture studio and have one of the largest teams of machine learning developers in the country. We can act as a technical co-founder and investor in developing a software solution.
This is the result of having the opportunity to chat with a lot of very smart people and their feedback.
Special thanks to:
Yoni Yechtman, Slow Ventures
Luke Sophinos, CourseKey
Aiyappa Bollera, Toro
Eli Dukes, Verticalized
Noah Nagle, Inovia Capital
Isaac Souweine, Pender Ventures
Pedro Gascon, Tiny
Cory Janssen, AltaML
Sam Chen, AltaML
Craig Haney, AltaML
Gagan Kapoor, AltaML
Ashneil Bhatnagar, Frankel & Company