The founding PMM's first deliverable isn't a positioning doc

When I joined Trezor in September 2023 as the first product marketing manager the company had ever hired, the first thing I did was nothing.

I didn’t write a positioning doc or run a workshop. I made a list of every person I needed to talk to, and I started talking. Nobody puts that part in the job description, but it’s the part that decides whether anything you build later actually survives contact with the company.

Nobody knew what I was

Trezor had been shipping hardware wallets for a decade without a product marketing function, a positioning framework, or a messaging hierarchy. The company had built one of the most trusted brands in crypto on instinct, great products, and a community that genuinely loved the work. People had gotten very far without me, and it was important that I started by understanding why before I started telling anyone what was missing.

So my first sixty days were spent meeting product managers, designers, the marketing team, support, leadership. I wasn’t pitching anything. I was asking questions, mostly variations of “how does this actually work here,” and listening for the answers underneath the answers.

What I found was a company that made product marketing decisions on a “let’s do something” and “I feel” basis. The instincts were good. They had gotten Trezor this far. What they couldn’t do was scale, survive a handover, or stay coherent across ten teams shipping in parallel. The biggest gap was between product and marketing. They operated next to each other, occasionally waving across the room. Marketing didn’t always know what product was building. Product didn’t always know what marketing was saying.

The work wasn’t missing entirely. Before I arrived, the CCO had written the key selling points for the Trezor Safe 3 himself because he saw it needed doing and nobody else was going to. He was right. But it was one person’s instinct filling a gap that should have been a process, and that’s how it works at every company without product marketing: the work gets done, by whoever feels the urgency first. My job wasn’t to start from zero. It was to take what people like him had been doing on instinct and turn it into something the whole company could repeat.

Before I could do any of that, I had to explain what product marketing even was. Most people at the company had never worked with a PMM. I used a metaphor that ended up sticking: “positioning is the foundation of a house, messaging is the structure, copy is the paint and the curtains.” That metaphor did more for the function than any framework I introduced afterwards, because people could explain it to each other when I wasn’t in the room.

The trust thing

Nobody warned me about this part. The most important thing I shipped in my first year wasn’t a positioning doc or a messaging framework. It was trust, in two directions at once.

You have to become the product expert on the marketing team and the marketing expert on the product team. Both sides have to feel like you understand their world well enough to speak for it when they’re not around. The marketing team has to trust that you won’t promise things product can’t deliver. The product team has to trust that you won’t flatten their work into something they’re embarrassed to put their name on.

The first time a product manager pulled me aside to gently correct a claim I’d written, I almost took it as a setback. Looking back, it was the opposite. It meant they cared enough to read what I was writing and trusted me enough to tell me directly instead of escalating it. That’s the moment the function actually started to exist. Not when I published the first positioning doc. When the people closest to the product started treating me like I was on their side.

The proof

Trezor Safe 5 launched in June 2024 at BTC Prague. It was my first full hardware launch, and the one where everything I’d built either worked or didn’t.

In the months before, I had run positioning workshops with stakeholders from product, design, brand, and leadership. We worked through competitive alternatives, unique attributes, and the value those attributes deliver. I built a launch tier system so the company could stop treating every release the same way. I wrote the messaging docs. None of these frameworks were exotic. You can find a positioning template in five minutes. The work was earning the right to introduce them and making sure they got used after I left the room.

I knew it had worked before the keynote even started, because of how the weeks leading up to it felt different. The content team didn’t ask me what to write. They opened the messaging doc and pulled the lines themselves. The video team had stopped asking what the angle was. PR already knew the headline. The product page, the launch video, the ads, the press release, and the on-stage presentation all said the same thing, not because someone was policing it but because the story was written down somewhere everyone could find it.

On Reddit and X, the community fed back the exact narrative we had built months earlier. They were using our words. The proof of good positioning isn’t that the company repeats it. The proof is that customers do.

We’ve used the same process since for Trezor Safe 7, for the Trezor Expert onboarding service, and for dozens of feature launches in between. Same bones every time.

What founding actually looks like

If I had to compress a year of building a product marketing function into one sentence, it would be that the work isn’t the frameworks. The frameworks are everywhere. The work is earning the right to introduce one, getting the company to use it, and being useful enough in the meantime that people invite you back into the room.

You walk in. Nobody knows what you do. You spend a quarter learning, a quarter teaching, and a quarter proving, and if you’re lucky, somewhere in the second year you watch a launch happen without you needing to be in every meeting, and you realize the thing you came to build has started to build itself.

That’s the job. Make yourself useful enough that the function outlives the founder.