PM job descriptions list artefacts. Roadmaps, requirements, specs, OKRs, presentations. They make the role sound like a documentation job.
The artefacts are the visible output. The actual work is the unbillable middle — the conversations, the trade-offs, the judgment calls, the nudges that keep the team moving in the same direction. None of it shows up in a JD because none of it can be turned into a tidy bullet point.
If you only do the work in the JD, you'll be a mediocre PM. The role lives in the parts that aren't written down.
The visible work versus the actual work
A short version of the gap:
The JD says "write the roadmap". The actual job is choosing what's on it. The writing is twenty minutes. The choosing is thirty conversations.
The JD says "manage stakeholders". The actual job is maintaining alignment when their incentives pull in different directions. That's a relationship maintenance job, not a meeting-running job.
The JD says "define requirements". The actual job is closing the gap between what the stakeholder thinks they want and what the user actually needs, in a form an engineer can build from without follow-up. Three jobs, one bullet point.
The JD says "use data to inform decisions". The actual job is figuring out which data matters, what it's lying about, and what to do when there isn't enough of it yet. Most decisions in early product happen with incomplete data. The skill is making them anyway and being honest about the confidence interval.
The unbillable middle
Five things that show up in every effective PM's week and never on a JD:
Translating between functions. Engineering, design, marketing, sales, exec — all speak different dialects. The PM is the translator. A surprising amount of what looks like consensus on the team is the PM quietly making sure each function heard a version of the plan they could act on. When this work is missing, alignment looks fine in the meeting and falls apart in execution.
Defending focus. The roadmap is constantly under attack — small asks, urgent fires, stakeholder enthusiasm for the new shiny thing. The PM's job is to keep the team focused on what they agreed mattered, without becoming the obstructive person who says no to everything. That's a daily judgment call, not a policy.
Asking the question nobody asked. In every meeting there's an assumption nobody examined. "Are we sure users actually want this?" "What happens if this doesn't work?" "Do we know how to measure success?" The PM who asks these questions gets called annoying for a quarter and indispensable for the next five years.
Holding the long view. Engineers see the next sprint. Stakeholders see the next deadline. Designers see the next release. The PM is one of the few people on the team whose job includes thinking about how today's decisions interact with where the product needs to be in six months. Without that, every decision optimises for the short run and the product slowly becomes incoherent.
Catching the small problems before they're big. The integration that's slipping. The team dynamic that's straining. The support ticket pattern that just appeared. None of these are crises yet. All of them will be in two months if nobody acts. The PM who catches them early looks like they're not doing much. They're doing the most valuable work on the team.
Why this work goes unrecognised
It doesn't ship. The roadmap ships. The feature ships. The retro deck ships. The unbillable middle is the connective tissue that makes shipping possible — and the credit for shipping always goes to the visible artefacts and the people who built them.
PMs new to the role often try to make their work more visible by producing more artefacts. More documentation. More dashboards. More status updates. That makes the work visible at the cost of doing it. The artefact factory is busy. The product isn't getting better.
The teams that recognise good PM work are the ones with senior people who've seen it before. They know what's happening when an experienced PM is in the room versus when there's a gap, and they protect that work. The teams that haven't seen it can't tell the difference, and they end up rewarding the artefact factory.
What the role actually rewards
Good PMs don't get promoted for the JD. They get promoted because the team they were on shipped good products and worked well together — and at some point someone notices the pattern.
The work that produces that pattern isn't on the bullet points. It's the listening, the translating, the focusing, the questioning, and the early catches. Do those well and the artefacts mostly take care of themselves.
If you're doing the fundamentals well, most of the unbillable middle is just the consistent practice of those fundamentals. And active listening is the skill that powers most of the work that doesn't make it into the JD.