The Punk PM #56

Hey there, punk!

We hosted ProductTank Newcastle again this week, with a cracking talk from former hedgehog lab CEO—and now solo founder-builder—Sarat Pediredla on running thredspan as a one-person product team.

The bit that stuck with me was about restraint. In a world where code is cheap and you can ship anything you want, the hard part isn't building. It's knowing what not to.

Let's get into it.

Quote of the Week 🙊

Insight 🦉

Anyone who grew up in the UK in the eighties remembers the Grange Hill kids telling us to "Just Say No". A heroin storyline in a children's drama, spun off into a charity single so wholesome it got a boatload of British child actors shipped to the White House to shake hands with Nancy Reagan. It was a terrible record. Of its time. Riding the moral panic of Nancy's husband Ron's forever-war on drugs.

The slogan was rubbish in 1986. But it's the best advice a product manager will get in 2026.

Because for us, the drug is shipping. Nothing beats the hit of getting a feature you specced into your users' hands. It's seductive. It's exciting. It's genuinely cool to watch something you imagined turn into working software. And here's the thing: AI has just turned a trickle into a firehose. When your agent can spec, design, build and ship in an afternoon, every "yes" feels free. CEO has a brainwave? Build it. Sales lost a deal over a missing toggle? Build it. Loudest customer in the room wants their pet feature? Build it.

Like cheap cocaine flooding Tony Montana's Miami, the supply has never been higher or the price lower.

But cheap to build was never the constraint that mattered. The cost was never the code. The cost is what each new thing does to everything around it. Every feature you bolt on dilutes the core. Blurs the message. Adds a setting someone has to maintain, a path someone has to test, a decision someone has to make on the way to the thing they actually came for. You don't notice it happening. You just wake up one day shipping a worse product, faster.

In 1997 Oasis released "Be Here Now". The follow-up to "(What's The Story) Morning Glory", and an absolute monument to overindulgence—seven, eight-minute songs buried under layer upon layer of guitar. Noel Gallagher has since admitted he lost the plot. What he needed wasn't more cocaine and more amps. He needed one person in the room willing to say that song doesn't need to be nine minutes. Three and a half is fine. You don't need another guitar on there. That person never showed up. The record is the sound of nobody saying no.

That's the job now. Not the building—the deciding what's worth building in the first place. Our role isn't to say yes to everything the AI can suddenly make cheap. It's to edit. To make sure what ships delivers the value we agreed to deliver, and does nothing more than it needs to in order to deliver it.

An editor doesn't prove their worth by the words they let through. They prove it by the ones they cut.

The alternative is Homer Simpson's car. Homer got handed a blank cheque and built the vehicle of his dreams—three horns, shag carpet, a separate soundproof bubble dome for the kids. It bankrupted the company and nobody wanted it. That's the trajectory of build-build-ship-ship-ship in the AI age: products that once did one thing brilliantly, customised into incoherence until they do nothing anyone needs.

So as software gets cheaper to make, brace for the flood. More products. More features. More mediocrity, churned out faster than ever, because saying yes has never been easier and saying no has never felt more like leaving value on the table. But the real value is in the no. The taste to see the core, and the nerve to defend it against every shiny, cheap, build-it-by-lunchtime idea that wants a piece.

The Grange Hill kids were wrong about a lot. But they had the chorus right. As the supply gets cheaper and the buzz gets stronger, the product managers who matter will be the ones who can look at the firehose, look at the dealer, look at their own itching hands—and just say no.

Action 🚀

Next week, look at your roadmap and find the feature you're least sure about. The one that's there because someone asked, not because you're convinced it earns its place.

Now make the case for cutting it. What does the product gain by not building it? What gets clearer, simpler, faster?

If you can't build a stronger argument for killing it than for shipping it, fine—build it. But make sure it earns that yes.

Inspiration 💡

Why Companies Should Have Product Editors, Not Product Managers – An old article from Andrew Chen on Square's practice of calling its PMs "product editors", and Jack Dorsey's line that editing is every leader’s real job—taking endless inputs and cutting them down into one cohesive story. The sharpest takeaway here: a bad idea is often a good idea that simply doesn't fit. That's the whole discipline of this week's Insight, made years before AI made shipping almost free. Read more

Chief Question Officer – Noah Weiss on the questions great product leaders use to interrogate a proposal. Two earn their place this week: "are there any cheat codes?"—what scope can you cut and still win—and "what are the anti-goals?"—what are you deliberately choosing not to do. Most roadmaps obsess over the goals. The harder, more revealing work is naming what you'll refuse to build. Read more

All My Clients Wanted a Carousel, Now It's an AI Chatbot! – Adële, a freelance web developer, on the depressing conveyor belt of features clients demand purely because a competitor has them. Carousels, then cookie banners, now inevitably a chatbot that confidently gives out the wrong opening hours. Nobody actually uses these things; they're social signals, not tools. Her closing line is the one to sit with: building something genuinely simple is harder than bolting on a chatbot, but nobody sees the restraint. 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