Software alone isn't a moat anymore

AI has made building stuff faster, cheaper, and far more iterative than before. If an enterprise wants to build their own software for their own specific needs, and they've got the expertise and the money, they can definitely do it in-house in 2026.

That doesn't mean building well is easy — it isn't. But the barrier to getting something built has dropped enough that software alone isn't a moat anymore.

The focus has shifted to whether your product delivers the outcome it promises — and whether you are the one who owns that delivery.

I've seen this operating system before — in professional services. Professional services firms have always been paid for outcomes, not for handing over artifacts or files to the client. That constraint forced them to create a different operating system — sit with the client to deeply understand their problem, design a solution specific to their needs, work across teams to deliver it, and stay accountable for the end outcome. Your job doesn't end when the invoice is signed. That's when it begins.

Building the software is turning out to be the easier part. Making it work inside one specific, often messy, real organisation, and proving it delivered value — is the harder problem now. I watched that discipline get built for years, on the client-facing side of professional services, and I'm starting to see the same instincts show up inside AI-native companies selling into enterprises.

The best client-facing operators I've worked with went much beyond just delivering great presentations or closing big deals. They were the ones who knew the client's business almost as well as they knew their own, and who translated hard and sometimes vague problems into a working solution. They owned the relationship, P&L, and the outcome — so when the delivery failed, it instantly showed up in their numbers.

An outcome-focused services mindset is closer to a discipline. You learn it by being the one on the hook, repeatedly.

AI is creating a new kind of company now — AI-native at the core, with human experts operating at the edges. The org chart is leaner and the tools are new. But the job hasn't changed. Someone still has to sit with the client, understand their problem, deploy solutions, and oversee final value delivery.

The shift isn't from services to product or vice versa. It's from selling the promise to owning the outcome — with better tools than we had before.

—Tushar