The Projects You Never Hear About
A project gets scoped in a conversation you weren't in. It's a rewrite of the scheduling service, the one you've been patching around for two years and have opinions about. Two names come up. Neither is yours.
By the time it lands on a board it's staffed and moving, and you find out when the kickoff invite arrives.
Sometimes that's ordinary staffing. Sometimes somebody pictured what the project would be like with you on it and picked around you.
You will almost never learn which one happened.
Being difficult to work with has a narrower effect than it seems. It leaves your assigned work alone, which is why it can run for years without anything visibly breaking. What it reduces is how often people offer you something, and the offered work is the ambiguous, unscoped, high-leverage kind. That's the work where growth actually lives.
The reduction leaves nothing behind. No document where you were considered and passed over. No thread you can go back and read. Your week fills up with other things, your throughput looks fine, and the only thing that changed is what you're being asked to do.
Routed work and chosen work
Routed work comes off the board, out of the rotation, through on-call, or from the service with your name on it. It arrives on a schedule that has nothing to do with how anyone feels about working with you. It came last month and it'll come next month.
Chosen work comes from a person deciding to bring you in. The redesign nobody has scoped. The problem that spans three teams and has no owner. The early conversation about what to build, before there's anything to build.
That second kind is where you get to exercise judgment, because it's ambiguous. Where you define the shape of something, because it's unscoped. Where people form opinions about how you think, because it happens in front of them.
Nobody is required to include you.
So the two channels can move in opposite directions for a long time. Your throughput holds because routed work has no opinion about you. Your scope flattens because the discretionary work went somewhere else.
What someone pictures before they add your name
When a lead decides who to pull into an ambiguous project, they're not reviewing your record. They're imagining the next six weeks.
- The design review. Does it take one round, or four, because every comment turns into something that has to be won instead of resolved?
- The decision meeting. Do they need a side conversation with you beforehand so the decision survives contact?
- Week three. The team settles something while you're heads down, and you come back and reopen it.
|
|
Where can AI save you time?
My friends at Big Creek Growth put together a quick survey to spot the repetitive work you can hand off to automation.
|
|
None of that shows up in a performance conversation. All of it shows up in how much of their week they're signing up for.
Casciaro and Lobo studied roughly 10,000 working relationships across five organizations and found that when people need help getting something done, they reliably pick the more congenial colleague over the more capable one. The research is twenty years old and it still describes what you can watch happen in any staffing conversation.
Nobody runs this as a calculation. No one sits down and scores you. They remember what the last project felt like, they pick accordingly, and they'd have a hard time explaining why if you asked.
The review names a trait instead of a cost
Your assigned work keeps shipping, so nothing in the system flags a problem. Velocity looks fine. Tickets close. No metric captures the projects you were never on.
When feedback does come, it comes late and soft. Something about collaboration, or communication style, or being more open to input. It names a trait instead of a consequence, which makes it easy to hear as a matter of taste and easy to set down.
Some of that softness is structural. Zenger and Folkman surveyed more than 7,600 people and found 44 percent said giving negative feedback was stressful or difficult. Telling a strong performer that people would rather not work with them is one of the hardest versions of that conversation. The vague version costs almost nothing to deliver.
I've written before about why careers stall even when the work stays good. That piece looked at engineers who kept executing well while expectations quietly shifted toward shaping work instead of delivering it. This is a second cause with the same symptoms. Same plateau, same silence, different mechanism, and the fixes aren't interchangeable.
The people who are difficult and get invited anyway
You've worked with someone who's genuinely hard to deal with and gets pulled into everything regardless. That person is real, and any version of this argument that pretends otherwise isn't worth reading.
What's happening there is a trade. They know something scarce enough that the friction is worth absorbing. The payments system nobody else understands. Ten years of context on why the architecture looks the way it does. When the alternative is shipping without them, the cost gets paid.
That trade holds until one of two things happens.
Someone else learns the thing. The scarcity was doing the work, and when it's gone, so is the tolerance.
Or the work changes shape. Deep individual knowledge matters less when a project needs four teams aligned. At that point the friction isn't just expensive. It threatens the outcome.
If you're the one absorbing that cost from someone else, raising the bar with a teammate directly is a better move than routing around them.
Count how much of last year you were chosen for
Self-assessment is useless here. Nobody thinks they're the difficult one, and asking yourself whether you're easy to work with returns an answer you decided on years ago.
Look at observable things instead.
- Sort last year's work into routed and chosen. What share arrived because a process sent it, and what share arrived because a person picked you? Then ask whether that ratio moved in either direction.
- Notice when decisions reach you. Are you consulted while something is still open, or told once it's settled? Being informed isn't the same as being asked.
- Watch for problems that aren't yours yet. When someone is stuck on something outside your area and wants a second opinion, do they come to you? That's the clearest read on whether people want you in a room.
Then there's the direct version, which is harder and worth more. Pick one person and one recent project you weren't part of.
What would have been different if I'd been on this?
That question is answerable. "How am I doing" isn't, and it gives you reassurance instead of information.
You might not like what you hear. That's the point of asking.
The behaviors that make people hesitate are usually small and specific. How you handle disagreement in a review. Whether a settled decision stays settled. How much interpretation your feedback needs before someone can act on it. Those are habits, and habits move in a way that personality doesn't.
Routed work will keep arriving either way. The ratio is the part that tells you something, and it's worth checking more than once a year.