Bringing a New Person Up the Curve

Category: Systems That Assume Reality A system is only as good as the speed at which its newest member can run it.

Toyota’s production system is famous for the andon cord and just-in-time delivery, but the quietly remarkable thing is how it brings a new worker on. A new person is not handed a binder and left to sink. They are placed beside an experienced worker and given a small, well-defined slice of the work — a single station, a handful of standardised steps written in plain terms, with the reason for each one attached.

The standardised work sheet at the station is not a bureaucratic relic. It is the curriculum. The new worker runs that slice, slowly, while the experienced one watches and corrects in real time. The pace and the scope widen only as competence is shown, never before. And the new worker is encouraged — required — to pull the cord when something is wrong, on day one, before they could possibly know better, because the system treats a newcomer’s confusion as information about the work and not as a deficiency in the newcomer.

Within weeks a person off the street is running real production at real quality. The achievement is not that Toyota hires unusually capable people. It hires ordinary ones. The achievement is that the system is built to climb — legible, sequenced, paired, and self-correcting — so the climb is fast and the result is reliable almost regardless of who starts it. The newcomer got good fast because the system was built to make newcomers good fast. The speed was never really about the newcomer at all.

The Principle

A good system is legible to the person who joined last, and it transmits itself on purpose. Onboarding is not a courtesy you extend to a newcomer. It is the test of whether your system is real, and the mechanism by which it survives the people who currently run it.

Two claims define the chapter, and both cut against the obvious reading. The first is legibility. A good system is understandable to the person who joined last, not just to the veterans who built it. The test of a system’s clarity is not whether an expert can run it — experts can run almost anything, including a fog they have privately memorised. The test is whether a stranger can see how it works. The second claim is deliberate transmission. A good system hands itself over on purpose, through sequenced tasks, a named guide, the why attached to the what, and live correction. It does not rely on osmosis, on the nearest desk, and on the new person’s nerve.

The comfortable misreading is that the climb is about the recruit. Are they up to speed yet? Are they a slow learner? Did we hire well? The turn is that the speed and quality of the climb are mostly a property of the system, not the person. A legible, self-transmitting system makes ordinary people useful fast. An illegible one makes even strong people flounder for months, and then quietly credits or blames them for a result the system produced.

So onboarding is not a soft nicety. It is a hard structural property. The rate at which a system can absorb a new person is the rate at which the organisation can grow, survive turnover, and find out whether what it built is real or only a habit a few people happen to share.

Why It Is Inevitable

A system that intends to outlast its current staff is pushed toward legibility and deliberate transmission, because the alternative fails in a way that stays hidden until it is expensive.

People leave and people join, always. Turnover is not a defect to be engineered out. It is the permanent weather. A system that can only be run by the people who happened to build it has a half-life measured by their tenure, and that clock is always running whether or not anyone is watching it.

An illegible system hides its illegibility behind its veterans. A team of long-tenured people makes an unlearnable system look fine, in just the way a sufficiently good expert papers over the absence of a written process and so deepens the dependency on themselves. Everything seems to work, right up until you try to add someone. Then the months-long flounder reveals there was never really a learnable system at all — only a group who had each privately memorised a fog and called the overlap between their fogs “how we do things.”

And growth forces the issue. You cannot scale a thing only its founders understand. The moment you need the eleventh person doing what the first ten do, the climb itself becomes the bottleneck. So any organisation that grows, or that simply persists across normal turnover, is driven sooner or later — gracefully or in a panic — to make its system climbable. The disciplined ones do it on calm days, before the painful hire, on purpose.

The obvious objection is that good onboarding is slow, expensive work that never beats the deadline, and the new person will figure it out eventually anyway. But “eventually” is the cost. It is months of half-productivity, repeated missteps, and a newcomer who learns the system wrong and then teaches the next one wrong — because what gets transmitted by osmosis is whatever happened to be nearest, not what is correct.

How It Shows Up

• A new starter is given a real but bounded task in week one, not a fortnight of reading and “settling in” — and the task is sized to be winnable.
• There is a named person whose explicit job is to answer the newcomer’s questions, so learning happens by deliberate pairing rather than by interrupting whoever is nearest and least busy.
• The written account a newcomer is handed carries the why, not just the steps, so they can reason past the first situation the document didn’t anticipate.
• Difficulty is graded on purpose: the scope and pace of real work widen as the newcomer copes, rather than the full load landing on day one or trickling in by accident.
• The newcomer’s confusion is treated as information about the system — “why is this unclear?” — not as a deficiency in the newcomer.
• “How long until a new person is useful here?” is a question the team can answer with a number, and the number is weeks, not “it depends, a year-ish, some never really get it.”
• The questions a newcomer asks in their first month quietly improve the system, because someone writes the answers down where the next newcomer will find them.

Why It Causes Benefit

When a system is built to be climbed, it gives back three things, and they compound.

The first gift is speed to value. A new person becomes genuinely productive in weeks, not quarters. The organisation gets back what it pays for sooner, and the team stops carrying a passenger for months — which is itself a kindness to the newcomer, who would far rather be useful than lost. Being slow and adrift is nobody’s preference; the climbable system spares people the experience of it.

The second gift is that the system gets a free audit on every hire. Each newcomer’s stumbles are an X-ray of where the system is illegible, and an organisation that listens fixes the murk for everyone, not just the new starter. The veterans had long since stopped seeing those gaps, because you cannot see what you have stopped noticing. The newcomer can see nothing else, which is exactly why they are valuable. The thing that looks like the newcomer’s weakness is in fact the system’s most honest feedback, available for one short window and then gone the moment they acclimatise.

The third gift is resilience and growth. A system people can be brought up the curve on quickly is a system that survives turnover, absorbs growth, and never lets competence pool dangerously in a few long-tenured heads. It is the circulation that keeps the body alive — knowledge flowing in at the same rate it flows out. And it compounds, because every well-onboarded person can onboard the next one. Good transmission breeds good transmission, so the climb gets shorter over time rather than being re-improvised, painfully, with each arrival. The same discipline that brings a newcomer up the curve is the one that lets a leaver go without the roof falling in.

How To Cultivate It

• Give a real, bounded task in week one. Not reading, not shadowing in the abstract — a genuine slice of work, small enough to win and real enough to matter. Competence is built by doing under guidance, and the early win tells the newcomer the system is climbable.
• Name a guide, and make it their job. Assign one person explicitly responsible for the newcomer’s questions, so learning is a deliberate relationship and not a tax levied at random on whoever is nearest. Osmosis transmits whatever is closest; a guide transmits what is correct.
• Sequence the climb and grade the difficulty. Lay out the order in which a newcomer should meet the work, widening scope and pace as they cope. A system that drops the full load on day one is not testing the newcomer. It is admitting it never thought about the climb.
• Transmit the why, not just the what. The reasons are the part that lets a person reason past the unexpected. Onboarding that teaches steps produces operators who freeze when reality leaves the script; onboarding that teaches reasons produces operators who can think.
• Treat the newcomer’s confusion as a defect report on the system. When a new person is lost, ask first why this is unclear, before asking why they don’t get it. Capture the answer, fix the murk, and the next person climbs faster. The first month is the cheapest legibility audit you will ever get.
• Make the newcomer improve the onboarding. Have them write down what confused them and how it was resolved, for the next person. This turns each climb into a contribution rather than a cost, and keeps the path up the curve alive instead of letting it rot into a dead binder.

What Good Looks Like

A healthy organisation here is not the one with the thickest induction pack and the warmest welcome lunch. It is the one with a system built to be climbed — legible to the newest person, sequenced so the climb is fast, paired so confusion is met by a person and not a wall, and self-correcting so every newcomer leaves the path clearer than they found it. The new starter is useful in weeks and never privately written off as slow, because everyone understands the climb is the system’s job and not the recruit’s burden.

The test of a system was never whether its veterans can run it. It is whether the person who joined on Monday can. A system only its builders understand is not a system at all. It is a habit they happen to share, and it dies with them. The well-onboarded newcomer is the living proof that the organisation built something real — something that exists outside the heads that made it, and can be handed to the next pair of hands on purpose.

A Reflective Question

Think of the last person who joined your team. How long did it take them to become genuinely useful — and when they were slow, did you treat it as a fact about them, or as a question your system had just been asked and failed to answer?