The Mid-Career Plateau — A Flattening Growth Curve Is Normal

Why It Feels Like You Stopped, and Why Now
A Flattening Curve Is Not a Failure
The Point Where the Standard Quietly Changes
Depth or Breadth — The Vertical Stroke in One Sentence
The Pressure of the Manager Track
The Real Risk in This Stretch Is Isolation
Further Reading

Why It Feels Like You Stopped, and Why Now

The work is fine. The system you own runs without much drama, and when a new requirement lands you can mostly see how it will be handled. Then one day the thought arrives: what is different about me compared with last year.

This feeling shows up especially often somewhere between year five and year ten. And the stagnation of this stretch is usually two things mixed together. One is growth that has actually stopped. The other is growth whose type has changed, so it no longer registers. Fail to separate them and you will prescribe the wrong treatment.

A Flattening Curve Is Not a Failure

Early growth had tick marks. Yesterday you did not know a piece of syntax and today you do; last month you could not run the deploy and this month you can; you pick up one unfamiliar tool at a time. Growth that consists of learning things you did not know is countable.

Growth after year five is a different kind. It is getting more accurate at choosing among things you already know. In the same design meeting you make a better call than you would have three years ago, and you abandon on day one a direction that would once have burned three weeks. This improvement is real, and it has no tick marks, because the mistakes you did not make leave no record.

So growth in this stretch is almost always underestimated. There is one way to check. Write down what the version of you from three years ago would have done in this situation. Usually a substantial gap appears, and that gap is the part that accumulated without tick marks.

One honest addition, though. A flattening curve and a flat curve are different. If your judgment three years ago and your judgment now really are the same, that is not growth of a changed kind; that is growth that stopped. The question above is a diagnostic, not a comfort.

The Point Where the Standard Quietly Changes

A large share of the confusion in this stretch comes from a transition nobody announces.

In the early years you are evaluated on what you made. How many features you shipped, how clean the code is. Then at some point the standard shifts from what you made to the change you made. How this design altered the team's next six months, how many meetings the document you wrote removed, whether the name you coined became the organization's vocabulary.

The transition is rarely communicated explicitly. So you get a situation where you are working hard against the old standard and the evaluation does not follow. That mismatch — making more and being recognized less — is exactly this.

The response is to change the standard yourself. Each quarter, ask this. Of the things I made last quarter, which would have happened roughly the same way without me, and which were different because of me. If the second list is empty, that emptiness is the substance of the plateau.

Depth or Breadth — The Vertical Stroke in One Sentence

One question always surfaces in this stretch. Dig deep into one field, or spread across several.

The question is placed slightly wrong. What you actually need is not one of the two but a single deep axis plus enough breadth to connect that axis to other things. Breadth alone leaves you with decision rights nowhere; depth alone narrows the places where the depth gets used.

So it is better to rewrite the question this way. Can I say in one sentence what my vertical stroke is. Not the name of a language or a framework, but the kind of problem you are the person for. Consistency in large volumes of data, operating latency-sensitive systems, moving legacy without cutting it off, designing interfaces where several teams are entangled.

If that sentence will not come, the substance of the plateau is not skill but focus. And focus is not something you build in a few months; it is something you assemble by aligning your next choices one at a time.

The Pressure of the Manager Track

In this stretch a management offer arrives, or the pressure is felt even when no offer comes, because plenty of organizations still have exactly one promotion path.

To keep the decision out of the hands of your mood, three checks are worth running.

Does handling people problems create energy or drain it. Recall the state you were in on the evening after a difficult conversation. This answer is not about taste but about stamina, and it governs several years.
Can you picture yourself two years from now writing almost no code. Plans to stay hands-on while managing mostly collapse in the first quarter. You have to ask in advance whether that state is acceptable.
Does an individual contributor track actually exist in this organization. A name on the level chart and a person occupying that level are different things. Without the second, the track exists on paper.

Then look at whether the decision is reversible. Does this organization have a path back from manager to individual contributor, and if it does, has anyone actually taken it. Reversible decisions can be made quickly; hard-to-reverse decisions deserve more time. The framework for that judgment is covered further in How to Decide Well.

The Real Risk in This Stretch Is Isolation

This last part matters most. Between year five and year ten, what actually erodes a career is not decaying skill but isolation. And isolation proceeds quietly.

It arrives in three forms. First, knowing only one organization's way of doing things. The longer you stay, the more that company's method feels like the industry standard, when it is often a practice that exists only there. Second, everyone who knows you being inside the company. That is also the problem covered in Part 4. Third, the feedback vacuum. As seniority rises, nobody reviews your code seriously and fewer people argue with your designs. It looks comfortable, and it is the most common route to growth stopping.

The prescription is modest. Once a quarter, talk shop with someone at another company. Explicitly ask for review from someone better than you in that area. And use your one-on-ones as an information channel rather than a status report. That last item is covered in detail in How Not to Waste a One-on-One.

Plateaus usually pass. They do not pass on their own, though; they pass when new input arrives. What this stretch needs is not more effort but friction from outside.