Mental model

Experience has a half-life. The skill is knowing which half is decaying.

Not all hard-won knowledge ages the same way. Some of it lasts a whole career — how people behave, how to read a trade-off, how to break a hard problem apart. Some of it is stale within a year — the syntax, the tooling, the "right" way to do things in a world that no longer exists. A project built five years ago ran on different technology, different costs, and a different ecosystem than you'd reach for today. This tool is a way to tell the two apart — so experts don't get trapped by their own past, and newcomers know when to listen and when to forge a new path.

01 — The half-life

Knowledge decays — but not at one rate.

Like a radioactive isotope, a piece of what you know loses half its value over some span of time. That span — its half-life — is wildly different for different kinds of knowledge. Drag the slider to move through the years and watch the four tracks separate.

5 yrs

02 — The spectrum

A map of what lasts — and what doesn't.

Lay the things you might "know" along a single axis: how long until half of it is out of date. On the left, knowledge with a shelf life measured in months. On the right, knowledge that barely moves in a career. Tap any point to see why it sits where it does.

◀ Months A few years A whole career ▶

Tap a dot to read it.

Pick a point

What decays fast, and what endures

The fast-decaying things are almost always specifics — this tool, this version, this year's best practice. The durable things rest on something that doesn't change much: human nature, incentives, maths, the physics of systems. Tap any dot above.

03 — Sort it yourself

Practice the judgement.

Here's a stack of things people say with the confidence of experience. For each one, decide: is this durable wisdom that still holds, or a dated specific that the world has moved past? The skill you're building is the sort itself.

Card 1 of 10 Score 0
0 / 10

04 — Two traps

Both ends of the ladder misjudge the half-life.

Get the half-life wrong in one direction and you cling to something that expired. Get it wrong in the other and you throw away something that still works. Each has a name, a set of tells, and a fix.

If you have the experience

The expertise trap

You treat a short-half-life fact as if it were a permanent truth. It was true — on the technology and economics of the time you learned it. The lesson outlived the conditions that made it true.

Tells

  • "We tried that in 2016 and it didn't work." (The thing that made it fail is gone.)
  • Your rules of thumb assume yesterday's costs, scale, and tooling.
  • You're arguing from a war story, not from the problem in front of you.
  • You haven't actually built anything in this space recently.
The fix: before you rule something out, ask what has to be true for this to fail — then check whether that's still true today. Re-run the cheap experiment. Update on the world, not the memory.
If you're early in the journey

The reinvention trap

You treat a long-half-life truth as if it were a fad, because it came wrapped in old packaging. So you rediscover — painfully — something people solved decades ago, and call it new.

Tells

  • "That's how they did it in the old days" is your whole argument against it.
  • You're rebuilding a hard-won lesson from scratch, at full cost.
  • New tools have you convinced the fundamentals changed too. (They rarely do.)
  • You skipped the "why did they do it that way?" question entirely.
The fix: separate the principle from the packaging. Ask an elder not "what did you do?" but "what did you learn — and what made it true?" Keep the durable half; forge your own path on the rest.
Notice they're the same mistake in mirror image: confusing the shelf life of a piece of knowledge with the confidence it's delivered with. Confidence isn't a half-life. The question is never "who's older?" — it's "how fast does this particular thing decay?"
05 — Listen or forge

The test: listen closely, or forge your own path?

You've got a specific call in front of you and a piece of experience pointing one way. Should you defer to it, or trust that the ground has shifted? Answer for the decision you're actually weighing — the needle moves as you go.

◀ Forge a new path Trust the experience ▶

Answer a few above

Most real calls land in the middle: keep the durable principle, and forge your own path on the specifics the world has changed.

06 — Go deeper

The half-life is a lens. Using it well is a habit.

The hard part isn't the idea — it's applying it in the moment, on your own experience, when your confidence is loudest. Teams that do this well move fast without relearning old lessons, and respect their veterans without being anchored by them. That's the work we do with the leaders and teams we coach: turning hard-won experience into judgement that keeps compounding.

Talk to us about coaching
TAKE ONE

Sort before you defer

When experience speaks, ask first: is this a durable principle or a dated specific? Defer to the principle; verify the specific against today.

TAKE TWO

Confidence isn't a half-life

How sure someone sounds tells you nothing about how fast their knowledge decays. Judge the claim by the thing it rests on, not the delivery.

TAKE THREE

Keep the durable half compounding

Invest most in what lasts — judgement, systems thinking, how to learn. Those curves barely decay, so every year on them stacks.