Will product marketing survive AI?

The role survives. The people whose value rested on producing the artifacts may not. Positioning documents, data sheets, competitive analysis, and sales briefings are all much faster to produce now, so the work that used to justify the headcount no longer does. What remains is the part that was always the actual job, which is getting an organization to agree on a story and act on it together.

What part of product marketing can AI not do?

Leadership. Building the assets, designing the strategy, and coordinating the go-to-market are all inputs. The job is convincing an entire company that the direction is right and getting them to move together behind it. A model can draft the document. It cannot walk into a room and change what a group of people decide to do.

Should product marketers stop executing and focus only on strategy?

Mostly no, despite that being the common advice. The best product marketers tend to be strong executors, because proximity to the problem, the customer, and the code is where knowledge comes from, and knowledge is what earns you the standing to lead. A strategist with no contact with the work has opinions rather than authority.

Do product marketers need to learn to code?

They need to think like developers, which matters more than shipping production software. The new generation of tools all expose MCP servers and command line interfaces, so they can be scripted and reshaped into whatever you need. The marketer who sees a repeated task as a process to automate gets very different leverage from the one who treats it as a program to staff.

What should a product marketer do first to stay relevant?

Get genuinely good at a coding agent, then convert your repeated processes into reusable skills rather than running them by hand each time. After that, keep experimenting, because the tools change weekly. Falling behind is less about missing a specific tool than about deciding the whole category is not for you.

Is PMM going to survive the AI reckoning?

For us as product marketers, the legitimate question is: will product marketing survive the AI reckoning?

And the truth is, the role will survive, but most of the people who are currently in it might not.

Why everyone is asking

The reason everyone's asking this is that they see all these layoffs. They see people who have been working at a company for 25 to 30 years being summarily dismissed, and they wonder when it's going to be their turn.

It's a legitimate concern.

I'd be lying if I didn't tell you that I feel this concern every single day. I'm in my mid-fifties, and I'm incredibly expendable. I'm also incredibly expensive.

And so if someone says I can replace that guy with an AI, it's probably true.

Except for the leadership question.

Leadership remains the fundamental PMM skill

It's not because AI is going to be great at product marketing. AI will probably not even be very good at product marketing.

As I've argued in the past, the essence of product marketing isn't writing data sheets or writing positioning frameworks or even doing customer research. The essence of product marketing is leadership. It's taking the assets that you've built and the strategy that you've designed and the go-to-market that you've coordinated, and convincing the entire organization that you're right and that they need to follow you and they need to work together. That will not change, and that is not something that an AI can do for you.

The standard advice that most people give is to stop executing and start strategizing. To some extent, that can be correct. I would say that the great product marketers are really great executors. It's because execution drives leadership. Proximity to the problem, customer, and code all translate into knowledge, which is power and which drives leadership.

What AI can actually do to you

But what an AI can do is make you obsolete and make you slower. The truth is, these new tools all have MCP servers and CLIs. They all can be coded and scripted. They all can be morphed into the tool that you've always wanted and that works the way you do.

So an old school product marketer is probably sitting around wondering what's going to happen to their job, whereas the new class of product marketer is actively learning how to code, learning how to use all these tools, and learning how to do in 15 minutes what used to take a week.

Think like a developer

It's not just asking questions in ChatGPT. It's learning how to think like a developer. Think in terms of processes and algorithms, and how you can improve those processes and algorithms continuously.

Don't just create a meetup program. Create a meetup process. Don't just create a webinar program. Create a webinar process that includes demand gen and activation.

What the job becomes

You still need to write positioning frameworks. You need to understand positioning at its core and how to find a unique and defensible stance in the market, especially a crowded market, especially a market that is under pressure from AI in all directions.

You still need to write data sheets.

You still need to write competitive analysis.

You still need to write sales briefings.

You still need to build customer journeys.

You still need to own messaging and segmentation.

But you just need to do all that stuff faster.

What to get good at

Earlier, I said, "Think like a developer." That's really it. Software engineers spend all day looking for ways to optimize processes and algorithms.

These are the three things you need to do today.

Get really good at Claude Code.
Turn your common processes into agent skills that you can use repeatedly.
Keep experimenting. Everything changes every week. You're not behind. You just need to get on the horse and ride.