The Punk PM #35

Hey there, punk!

I can’t believe it’s been a year since I relaunched The Punk PM.

I went into it with bold intentions of publishing every Friday, without fail. That didn’t quite happen. But I’m still really proud of the thirty-odd issues that made it out into the world. Each one was written because I had something I actually wanted to say, not because the calendar told me to.

That part matters to me.

Last year, I deliberately took August and December off from writing. No newsletters. No social media posts. No content treadmill. Just room for family, life, and a bit of headspace. I’m planning to do the same again this year. If I end up publishing forty-something issues in 2026, I’ll call that a win.

I never want this newsletter to feel performative. I don’t want to write for a schedule. I want to write when there’s a point to make.

So for the first issue of the year, it felt like a good moment to zoom out and take stock. AI hasn’t gone away. Product management is still shifting under our feet. And 2026 is shaping up to be a year where a lot of half-formed trends start to harden into reality.

Let’s talk about what that might mean.

Quote of the Week 🙊

Insight 🦉

If 2025 was the year of AI proof-of-concepts, 2026 is when things stop being cute and start getting real.

Last year was full of demos, prototypes, and “look what I built in a weekend” energy. Useful. Exciting. And, at times, a little distracting. This year feels different. The tools are bedding in. The behaviours are changing. And the implications for product managers are much bigger than another wave of features with “AI-powered” bolted on.

One of the clearest signals came at the tail end of last year with the rise of AI-native browsers and assistants. That’s not just a UX tweak. It’s a fundamental shift in how people interact with software.

We’re no longer designing purely for neatly contained web apps and smartphone screens. Increasingly, our products are being encountered through prompts, summaries, and conversations. Your user might never open your interface at all. They’ll meet your product inside a chatbot, a browser sidebar, or an agent acting on their behalf.

That changes the job.

It pushes us away from screens and flows, and towards intent, context, and outcomes. What does your product actually do for someone when they need it, regardless of where that interaction happens? If your value doesn’t survive abstraction, it’s in trouble.

At the same time, the longer arc we talked about last year is continuing. Product management is snapping back to its roots as a business function. ROI. Risk. Trade-offs. Financial value.

The ZIRP era hangover is finally wearing off. The days of huge teams optimising individual buttons, obsessing over MAU graphs while the company quietly haemorrhages cash are over. The correction back to what product management is really about will only accelerate in 2026.

What’s interesting, though, is the split emerging inside the discipline.

On one side, you’ve got customer-facing, growth-oriented PMs. Commercially sharp, close to revenue, accountable for results. On the other, a growing cohort working inward. Experimenting with AI-powered internal tools, automation, and bespoke infrastructure to drive productivity.

That shift also changes the SaaS story. We’ve talked a lot about the “death of SaaS” post-AI, and yes, that’s overstated. But something real is happening. Organisations are moving from buying tools to building them. When every company can spin up internal AI systems tailored to its workflows, the role of products inside large organisations starts to look very different.

It’s no longer about choosing vendors, it’s about shaping systems.

Meanwhile, prototyping tools have flattened the landscape. When anyone can generate a half-decent interface from a handful of prompts, everything starts to look the same. Same layouts. Same patterns. Same tasteful gradients. Competent, but boring.

That’s where things get interesting.

The real differentiation won’t come from doing what the tools make easy. It’ll come from pushing against it. We’re going to see the rise of what I’d call punk products. Opinionated. Human. Slightly weird. Products with a point of view.

And that demands courage. Taste. Judgement.

Which brings me to the quieter shift I think will define 2026.

We’re hitting the limits of our initial LLM obsession. The novelty is wearing off. The “AI everything” phase is giving way to a renewed focus on the human things machines still struggle with.

Meaning. Trust. Authenticity. Understanding how a product fits into someone’s actual life, not just their workflow.

In a world drowning in AI-generated slop, the products that stand out will be rooted in real human needs, real research, and real value. And the PMs who rise will be the ones who lean into that. Less time prompting machines. More time understanding people.

AI will still be everywhere in 2026.

But the edge won’t come from using it.

It’ll come from knowing when not to.

Action 🚀

Next week, practice not using AI.

Pick one decision, problem, or idea you’re working on and don’t reach for a tool straight away. Talk to a user. Sit with the mess. Write it out longhand. Let your judgement do the work first.

Then, and only then, bring AI in to help you move faster.

Inspiration 💡

12 Outlooks for the Future: 2026+ – Former Chief Strategy Officer at Adobe, Scott Belsky zooms out on what the next phase of AI change actually feels like. Less hype, more craft. As content floods the zone, proof of real work, shared experiences, and human taste become the signal. A sharp, optimistic read on where value really accrues next. Read more

Eight Predictions for Technology Consulting in 2026 – My ex-CEO Sarat Pediredla argues that consulting is being rewired from the inside out. Billable hours give way to outcomes, polish gives way to judgement, and AI-native firms blur the line between services and software. Required reading if you work in professional services, or anywhere near enterprise, platforms, and internal tooling. Read more

AI and the Human Condition – Stratechery’s Ben Thompson tackles the uncomfortable question behind all the AI noise: what happens to human value when machines can do almost everything? His answer is more grounded than the doomer takes. Even in a fully automated world, people still crave human-made things, meaning, and connection. A thoughtful reset. Read more

Signing Off ✍️

If this resonates with you, hit reply and let me know. And if you think a friend or colleague would enjoy The Punk PM, feel free to share it with them!

Play it your way,

Toby