Which SaaS Competitive Advantages Still Hold Up in the AI Era

For the last decade, building a defensible SaaS business followed a pretty proven playbook. Iterate until product market fit, get to scale, lock in your users, build a brand, and let the growth flywheel spin. It wasn't easy, and it took a little luck along the way. But if you pulled it off, the recipe was pretty tried and true.

Then AI showed up and seemingly blew up that playbook almost overnight.

Not all of it. But enough that founders and product leaders need to do an honest audit of where their competitive moats actually stand. Because some of the defenses you've been building are stronger than ever, and some are already full of holes.

One of the simplest frameworks to break this down is Hamilton Helmer's Seven Powers. It's the closest thing the strategy world has to a field guide for durable competitive advantage. Let's walk through each one, translate it out of the business-school buzzwords, and think through how AI might impact it.

Scale Economies

The basic idea: the bigger you get, the cheaper it gets to serve each customer. Your costs drop as your volume grows, which means you can either undercut competitors on price or invest more than they can afford to.

In SaaS, this showed up most clearly in your early up front investments. Once you've built the platform and absorbed the fixed costs, adding the next thousand customers is almost free. That's one of the foundations of the amazing valuations that growing ARR gives you.

In an AI world a lot of this is still valid, but there's a catch.

AI can actually amplify scale in some ways. The more data you have, the better your models get, which is a new kind of flywheel. But it also lowers the cost of that initial build for everyone. A two-person startup can now do things that used to require a team of twenty. The scale moat doesn't disappear, but the runway you need to build it has gotten a lot shorter.

Network Effects

This one is the holy grail and became especially evident with the rise of social networking. But the value of a network isn't exclusive to likes and shares. Every new user makes the product more valuable for every other user. Once Salesforce became entrenched, admins who knew the platform were likely to not just recommend it to peers, but advocate for it in their next job. Having your brand become a skill requirement in job descriptions was one of the best signs of strong PMF in the age of SaaS.

Even with AI this likely remains the strongest moat still standing.

AI doesn't erode network effects, if anything, it deepens them. If your product gets smarter from user behaviour and interaction data, then a larger network means a better AI, which attracts more users, which feeds a better AI. If you have genuine network effects, these days you need to protect them obsessively.

Counterpositioning

This is when a new entrant adopts a business model that incumbents can't copy without hurting themselves. The classic example is Netflix versus Blockbuster. Blockbuster couldn't go all-in on streaming without torching their own retail revenue. The moat was built into the heart of the business itself.

In SaaS, this plays out when a startup can take a stance the incumbent is structurally unable to match.

AI is creating new counterpositioning opportunities right now.

If you're a legacy vendor with a large services business, you can't fully embrace AI-led automation without undermining your own revenue model. Startups can, they get to start from scratch. That's a genuine counterpositioning window, and it won't stay open forever. If you're building in a space with fat, services-heavy incumbents, this is your moment.

Switching Costs

Once your software is embedded in a customer's workflows, data, and team habits, leaving becomes painful. It's not that your product is necessarily better anymore, it's that the cost of switching is too high to bother.

This has been one of the most reliable SaaS moats for a long time. Deep integrations, proprietary data formats, and workflow dependency all make your products sticky. There's a reason why some cynics will say that the most successful SaaS have hostages as much as they have customers.

This is where AI might be the most disruptive.

AI agents are getting very good at migration tasks. Things like extracting data, rebuilding workflows, bridging integrations used to be the foundation of high switching costs. Those are the same tasks that are right in the crosshairs of AI. They aren't gone yet, but you can almost see the writing on the wall. If switching costs are your primary moat, you need a second one.

Brand

Brand is trust at scale. It's the reason a customer picks you before they've had a chance to evaluate you properly. In SaaS, this often shows up as category ownership, the feeling that one company just is the answer to a particular problem.

AI hasn't really changed this, in fact it may be reinforcing more focus on the first principles of brand.

Brand used to be built on polish, marketing spend, and G2 reviews. In the AI era, it's increasingly built on authenticity and demonstrated expertise. Anyone can generate a professional-looking website and competent-sounding copy now. The brands that stand out are the ones where there's a real point of view behind the product. In a world drowning in AI-generated content, an authentic voice may be the most important moat of all.

Cornered Resources

This one is about exclusive access to something others can't get. A unique dataset, a key patent or a regulatory relationship could be an example of this.

In SaaS, proprietary data has been the most common version of this play. If your product generates data that nobody else has, that's a real advantage.

With AI this may be the most contested moat right now.

Every AI company is chasing proprietary data. The race to corner training data, user behaviour data, and domain-specific datasets is the defining business competition of this era. If you have a genuine data moat, it's worth more than it was two years ago. If you don't, you need a clear-eyed answer to the question of how you'll ever get one.

Process Power

This is the hardest one to explain, and arguably the most durable. It's when your organization has developed ways of working that are genuinely better. Not because of technology or data, but because of accumulated knowledge, operational discipline, and embedded culture. It's the difference between knowing what to do and knowing how to actually do it.

If you could hand a competitor your entire codebase, and they still couldn't catch you, it's because what makes you good isn't in the code.

In a world of AI this is still valid, and potentially underrated.

AI can replicate a lot of things. It cannot replicate organizational judgment, institutional learning, and the kind of culture that consistently makes good decisions under uncertainty. If your process power is real, AI can actually amplify it. You can move faster, learn faster, and execute better.

So where does that leave you?

If I'm advising a founder right now, here's how I'd think about things:

The moats that have gotten stronger in the AI era: network effects, cornered resources (if you have real data), brand built on genuine point of view, and counterpositioning against AI-averse incumbents.

The moats that are weakening: switching costs, scale economies at the low end, and any brand built primarily on polish rather than substance.

So the playbook hasn't been thrown out but you should be thoughtful about which pages you're spending your time on.