I get a hundred pitches a week. Ninety-nine of them are about the same thing: a new AI model that’s going to change the world. A new chatbot, a new image generator, a new something-or-other that’s marginally better than the last one. I get it. I’m an investor in Anthropic, OpenAI, and Scale AI. I love a good model. But I’m going to tell you a secret that will save your startup: the model is not the business.
The real war isn’t won on the leaderboard of the latest benchmark. It’s won in the trenches. It’s won in the plumbing, the wiring, the unglamorous, gritty, and often “boring” work of building infrastructure. If you’re in a vertical SaaS industry, your brand, your defensibility, and your legacy will be built on this foundation. Forget the hype. Let’s talk about what actually works.
My $30 Million “Boring” Exit
When we started RemoteTeam, the HR software market was about as exciting as watching paint dry. It was a stagnant sea of clunky, soul-crushing enterprise software that looked like it was designed in the 90s. We weren’t building a flashy social app. We were building tools for international payroll, time-off tracking, and local compliance. It was the definition of a “boring” business.
So how did we end up getting acquired by Gusto for a reported $30 million? We didn’t do it by adding more features than our competitors. We did it by building a brand around one simple promise: we make the excruciating pain of managing a global team disappear. That was it. We didn’t try to be an all-in-one HRIS. We focused on a very specific, very painful niche and we went a mile deep.
I remember one of our early customers, a fast-growing startup with engineers in five different countries. Their HR manager was spending a week every month just trying to figure out the different tax laws, payroll deductions, and statutory holidays. It was a nightmare. We built a system that automated all of it. A single dashboard to manage everything. That wasn’t a “feature.” It was a lifeline. Our brand wasn’t built on a clever marketing campaign; it was built on the back of a product that solved a real, expensive, and mind-numbingly tedious problem. We made the boring stuff simple. That was our brand.
The Real Moat is Buried in Your Infrastructure
This brings me back to the AI world. The models are getting commoditized faster than anyone expected. What was state-of-the-art six months ago is now an open-source project you can download for free. The real differentiator, the only defensible moat left, is the infrastructure that runs these models. The game is all about efficiency, cost, and speed.
I saw this firsthand a few months ago. We were scaling up a new AI-powered SaaS tool, and the cloud bills were starting to give me heart palpitations. The default answer from everyone on the team was “throw more GPUs at it.” It’s the easy answer. But it’s almost always the wrong one.
I had a hunch we were leaving millions on the table. So, I pulled two of my best engineers and we spent a solid month doing a deep, exhaustive bake-off: the latest and greatest GPUs versus Google’s TPUs. We didn’t just run some synthetic benchmarks. We took a massive, representative sample of our production inference workloads—a complex mix of language generation, code completion, and classification—and ran it across more than a dozen different hardware configurations. We measured everything: raw throughput, p99 latency, and most importantly, the cold, hard cost per million tokens.
The $2 Million Surprise
The results were staggering. For our specific workload, the latest-generation TPUs were not just a little bit cheaper. They were 40-60% cheaper. That’s not a rounding error. When I extrapolated the numbers, it translated to over $2 million in projected savings over the next 12 months. For a startup, that’s not just a nice-to-have. That’s your entire seed round. That’s the difference between scaling and dying.
It turned out that the parallel processing architecture of the TPUs was a much better fit for our model. We could handle a significantly higher number of concurrent requests on a single chip, which drove our costs through the floor. The GPUs were incredibly powerful, but we were paying for a Ferrari to go grocery shopping. We were wasting a ton of expensive horsepower.
This is what I mean by building a brand in a boring industry. While your competitors are raising massive rounds to pay their AWS bills and bragging about their A100 clusters, you can be quietly building an insurmountable competitive advantage by obsessively optimizing your infrastructure. Your brand becomes about being the smart, sustainable, and efficient choice. You can pass those savings to your customers, undercutting the competition and winning on value. That’s a story that resonates with CFOs, and in B2B, the CFO is often your real customer.
The Playbook for Boring Brilliance
So, how do you put this into practice? This isn’t just theory. This is a playbook I’ve used to build my own companies and one I look for in the 200+ startups I’ve invested in.
1. Go an Inch Wide and a Mile Deep
Stop trying to build the “everything platform.” It’s a trap. It leads to a mediocre product that doesn’t do anything particularly well. Pick a single, painful, and specific problem in a vertical you understand intimately. And then go all-in. Become the undisputed best in the world at solving that one problem.
At RemoteTeam, we could have easily been tempted to add features for performance reviews, recruiting, or employee engagement. But we said no. We stayed laser-focused on the unique challenges of remote work. That focus allowed us to build a product that was ten times better than the generic HR platforms for our target customer. Your brand becomes synonymous with that one thing. When people have that specific pain, your name should be the first one that comes to mind. That’s the power of focus.
2. Sweat the Small Stuff—It’s All That Matters
In “boring” B2B industries, the details are everything. A clunky user interface, a poorly documented API, or slow customer support can kill you. Your customers are betting their business on your product. They need to trust you. And trust is built in the small moments.
I once saw a multi-million dollar deal almost fall apart because of a poorly designed billing page. It was confusing, and it made the customer feel like they were being tricked. We spent a week redesigning it, making it crystal clear and transparent. It seemed like a small thing, but it saved the deal. That obsessive focus on the small stuff, on the entire customer experience, is what builds a brand that people trust. It shows you care. It shows you respect their business.
3. Build a Community of True Believers
Your customers in a vertical SaaS market are experts. They live and breathe their industry. They don’t want to be marketed to. They want a partner who speaks their language, understands their challenges, and is committed to their success. The best way to do that is to build a community.
We did this at MovieLaLa, a social platform for movie lovers that was acquired by Gfycat. We weren’t just an app; we were a hub for people who were as obsessed with film as we were. We hosted Q&As with up-and-coming directors, we sponsored local film festivals, and we created a space for genuine conversation. That community became our most powerful growth engine. They were our evangelists. They were our brand.
4. Turn Your Infrastructure into a Marketing Weapon
Don’t hide your engineering brilliance. Celebrate it. Your infrastructure choices, your performance benchmarks, your obsessive optimization—that’s not just a cost center. It’s a marketing asset. Write detailed blog posts about your journey from GPUs to TPUs. Share your benchmark data. Open-source some of your internal tools.
This does two things. First, it builds a powerful brand with your customers, who see you as the smartest, most efficient player in the market. Second, it helps you attract the best engineering talent. Great engineers want to solve hard problems. They want to work at a place that values technical excellence. Turning your infrastructure into a public part of your brand is one of the most powerful recruiting tools you have.
The Future is Boring
I’m more bullish on the opportunities in “boring” vertical SaaS than ever before. The AI revolution will reshape every industry on the planet, and the biggest winners won’t be the companies building the flashiest models. The winners will be the ones who are quietly, obsessively, and brilliantly building the essential infrastructure that powers this new world.
So if you’re an entrepreneur in a niche, unglamorous industry, my message is simple: you’re in the right place. You’re not missing the party; you’re building the house where the party is happening. Embrace the boring. Go deep. Sweat the details. And never, ever be afraid to get your hands dirty with the plumbing. That’s where you’ll find your moat. That’s where you’ll build a brand that lasts.
Frequently Asked Questions
How long does it take to build a brand in a boring vertical saas industry?
The timeline varies depending on your starting point and resources. For most founders, expect 2-4 weeks for initial setup and 2-3 months to see meaningful results. I've seen teams move faster when they focus on one thing at a time rather than trying to do everything at once.
What are the most common mistakes when building a brand in a boring vertical saas industry?
The biggest mistake I see is overcomplicating things early on. Start with the simplest version that works, get real feedback, and iterate from there. Another common trap is copying what worked for someone else without understanding the context behind their decisions.
How do I measure success with this approach?
Pick one or two metrics that directly tie to your goal and track them weekly. Vanity metrics like page views or follower counts rarely matter. Focus on metrics that reflect real engagement or revenue impact.
Do I need technical skills to build a brand in a boring vertical saas industry?
Not necessarily. While technical understanding helps, the most important skills are clear thinking and the ability to break problems into smaller pieces. Many successful founders I've invested in started with zero technical background and either learned enough to be dangerous or found the right technical partner.