The AI boom has made it easier than ever to ship software — and harder than ever to get anyone to care.
On the season finale of TechCrunch’s "Build Mode" podcast, Paul Irving, partner and COO at GTMfund, argues that most startups are still playing by an outdated sales and marketing rulebook. In his view, "distribution is the final moat in the AI era."
The old SaaS playbook is breaking
For years, the path was predictable: build an enterprise SaaS product, hire a sales team in a specific order, then scale it up with repeatable processes.
That one-size-fits-all approach, Irving says, "may have worked in the days of traditional enterprise SaaS but won’t cut it" in an AI-driven market where:
- Innovation cycles have compressed — what took years now happens in months.
- Dozens of competitors can ship similar features at roughly the same time.
- Buyers are flooded with tools that look and sound alike.
In that world, trying to win on product alone is a losing strategy. GTMfund’s thesis: the way you reach, educate and convert customers is now the real defensible edge.
Differentiate on distribution, not just product
Irving frames go-to-market as a series of highly specific choices, not a template to copy.
"The aperture for how you build your go-to-market or revenue engine and the decisions you make to get there have never had more unique and specific pathways depending on your company," he says.
Instead of chasing every channel from day one, GTMfund pushes founders to:
- Pick a narrow set of places where their buyers already hang out.
- Use AI and data to sharpen targeting and messaging.
- Delay heavyweight hiring until they understand what’s actually working.
A Facebook group as a primary channel
One example Irving shares: a startup that treated Facebook groups as a core distribution channel, not an afterthought.
By either starting its own groups or becoming "the most active posters" in a handful of highly relevant ones, the company put itself in front of exactly the right people. In one scenario, he says, you might have a group with 1,000 members where 700 are your actual ideal customer profile (ICP) and buyer.
If your goal is to land 40–60 new customers a year, that single channel "can become a really unique" growth engine, Irving notes. No paid ads. No huge sales headcount. Just sustained, focused participation where the right customers already gather.
What early-stage investors want to see instead
According to Irving, firms like GTMfund are not impressed by:
- Startups that burn half their budget on paid ads, or
- Teams that rush to hire a full sales org before they’ve found a repeatable motion.
They’re looking for founders who:
- Use AI to run smarter experiments with their go-to-market,
- Show creative traction in unexpected channels, and
- Understand precisely who they’re selling to — and why those people should care.
Don’t go it alone: build a real network
A constant refrain in the conversation is that founders shouldn’t try to solve distribution in isolation.
GTMfund leans heavily on a network of experienced operators and go-to-market leaders, but Irving says they don’t just hand over a giant contact list. Instead, they create bespoke introductions that are likely to produce real value for both sides.
He argues this collaborative spirit extends across the venture ecosystem:
"I think what’s wonderful about the venture-backed ecosystem more generally is people’s willingness to help," Irving says. "They know how hard it is to build a company. They know how low the likelihood of success is in the grand scheme of things, but people are always willing. If you come and you’re curious, you teach them something as part of that as well. People are usually willing to open up doors."
For AI founders, the message is blunt: shipping features isn’t the hard part anymore. The real work — and the real moat — is figuring out distinctive, data-driven ways to get those features into the hands of the right customers, then surrounding yourself with people who’ve done it before.
