Building Something That Outlasts Your Name on It

She built a process in her second year on the team, wrote it up in an afternoon, and taught it to four people. Six years later it had been renamed twice, was credited on the internal wiki to somebody she had never met, and was running in four departments including two that had never heard of her. She found this out by accident and spent a genuinely confusing week deciding whether she was annoyed or finished.

The correct answer, it turned out, was finished. Not resigned, not overlooked, finished, in the way a builder is finished when the building stands up without scaffolding. That is a much larger outcome than the alternative, which is a thing with her name welded to it that nobody could run without calling her first. This article is about designing deliberately for the first outcome, because almost every piece of advice available on the word legacy is quietly designing for the second one.

How do I tell whether my work will survive without me?

Take your name off it and see whether anybody could run it on Monday. If the answer depends on you being reachable, you have built a job rather than a thing, and jobs end.

Run the test literally. Pick the piece of work you are proudest of this year and ask three questions about it. Could a competent stranger execute it from what is written down? Are the judgment calls you make by instinct captured anywhere as rules? If you went quiet for a month, what stops, what degrades, and what carries on unchanged?

Most people find that the answer is uncomfortable in a specific way: the deliverable survives and the practice does not. The document lives on and the reason it was built that way lives only in your head. That gap is the entire subject of this piece, and it closes in about a page of writing.

Does it matter if I do not get the credit?

It matters less than it feels like it does, and the arithmetic explains why. Credit you have to defend is maintenance, and maintenance is a cost you pay forever.

Anything built to carry your name has to be protected: from renaming, from forking, from the person who describes it in a meeting without mentioning you. That protection costs attention every year for as long as the thing exists, and the attention comes out of the same account you were going to use for building the next one. Anything built to run without your name gets adopted, copied, improved and carried into buildings you will never enter, and it costs you nothing after the day you hand it over.

Kimberly Wade-Benzoni's research on intergenerational decisions, summarized in Current Opinion in Psychology, finds that legacy motivation is a real and surprisingly strong driver: people will give up present resources for the benefit of people who cannot reciprocate, when they can see that something of theirs persists. The interesting part for anybody ambitious is that this is not selflessness. It behaves like a form of long-range self-interest, and it produces more durable results than the version that keeps score.

How do I write it down so somebody else can run it?

One page, in the order a person would actually do it, with the three decisions you currently make by instinct written out as rules. The instinct part is the only hard part and it is the only part that matters.

Anybody can list the steps. What does not usually get written is why step four comes before step three, what you do when the input is malformed, and which of the rules you break and under what conditions. Write those. They are the difference between documentation nobody uses and a page somebody can actually run on a bad Tuesday.

There is a return on this that arrives immediately. Researchers reporting in Memory and Cognition found that people who studied material expecting to teach it later recalled more and organized it better than people expecting a test. Writing your work down for a successor puts you in exactly that state. You will find the two places you were faking it inside the first twenty minutes, which makes the page worth writing even if nobody ever reads it.

Who should I hand it to?

The person who has already been doing the awkward parts without being asked. Then say their name in the room where it gets decided, because a handover without sponsorship is not a gift, it is extra work landing on somebody with no authority to do it.

This is the practical half of legacy and it is a person, not a document. Handing something over has two moves in it and most people only make the first. The first is the transfer: the page, two sessions of doing it together, one session of watching them do it while you say nothing. The second is the public part, which is telling the room whose it is now, by name, in front of whoever controls their next opportunity. "Priya runs this now" said in a planning meeting is worth more to her than the six months of coaching, because it converts your credibility into her standing, permanently, at a cost to you of four words.

Do the second move even when it is inconvenient, and especially when the thing is good. The instinct in the moment is to stay attached to something successful for one more quarter. That quarter is exactly how a durable thing turns back into your personal job. Say the name, then get visibly out of the way, and let the next set of improvements happen without your approval on them.

What if handing it over makes me replaceable?

Being replaceable in your current job is the precondition for getting the next one. The person who cannot be moved because nobody else can run their work is not safe, they are stuck, and the cost of being stuck is paid in the rooms you never get invited into.

Adam Grant, writing in Harvard Business Review about givers and takers, describes a pattern that plays out over years rather than weeks: the people who spend their effort making others more capable tend to accumulate an unusual amount of goodwill and information, and it shows up late rather than early. The short-run scoreboard favors whoever hoards. The long-run one does not, because the room eventually fills up with people who know exactly what you did for them.

Design for that on purpose. Every piece of work you own should have a name attached that is not yours by the end of its second year, and you should be able to say what that name is. That is not humility. It is capacity planning for an ambitious decade.

The version that keeps running

The test is whether you could hand it over on a Friday. Not whether you would want to, not whether the timing is convenient, but whether the page exists, the person exists, and the room knows their name. If all three are true you have built something. If any of them is missing you have built a role, and roles do not outlast the person filling them.

None of this is an argument for wanting less of the credit. Take the credit while it is being handed out, put it on your review, say it out loud in interviews. The distinction is between credit you collect and credit you have to guard, and only one of those is free. The builder who ends a decade with six things running in other people's hands has a larger claim than the one who ends it with two things nobody else can touch, and has spare attention left over to start the seventh.

Pick the piece of work you are proudest of. Write the page this week. Then find the person who has been doing the awkward parts, and say their name in the room.

Sources

• Current Opinion in Psychology, Legacy motivations and the psychology of intergenerational decisions
• Memory and Cognition, Expecting to teach enhances learning and organization of knowledge in free recall of text passages
• Harvard Business Review, In the Company of Givers and Takers