Career Maintenance Is a Regime, Not a Reaction

September, and the reflex

Something happens every September. The year is two-thirds gone, the planning cycle starts, a reorg gets announced somewhere in the building, and people open a document they haven't looked at since the last time something like this happened. The résumé. It's dated eighteen months ago, sometimes longer, and it describes someone who had a slightly different job.

Then the weekend goes. Old performance reviews, if they were kept. Emails searched for a number that might have been $11 million or might have been $13 million. A project that took eighteen months compressed into a line that could have been written by anyone who was in the building at the time. By Sunday night there's a document. It is mostly accurate. It's also thinner than the work was, and the person who did the work is the only one who can tell.

That's the reaction. Most senior professionals have run it four or five times, and each time it feels like a one-off, because the trigger is always an event. The reorg. The recruiter. The January or September reflex. Nobody plans a reaction; that's what makes it one.

This post is the centre of a series that has been circling one idea since June: that a career record is a thing you maintain, not a thing you rebuild. The earlier posts took the idea one piece at a time. This one puts the pieces together and names what they make.

Why the reaction can't work, even when it's careful

Start with memory, because that's what the reaction runs on.

Ebbinghaus plotted the forgetting curve in 1885 and a University of Amsterdam team replicated it in 2015(opens in new tab) with the same shape: most of what you lose, you lose almost immediately, within the first day or two, and the slow decline after that never gives it back. What survives long-term is the gist. The outcome, the headline, the fact that the migration happened. What goes first is the texture: the constraint that forced the phased cutover, the design you argued down, why the timing mattered. Those are exactly the things a senior interview asks about and exactly the things a September weekend can't recover, because they were never written anywhere.

Then there's the shape of a career. The US Bureau of Labor Statistics(opens in new tab) puts median tenure with a current employer at 3.9 years, and even at 45 to 54 it's 7.0 years. Twenty years of work is spread across four or five employers, and more if you've done contract work. No single organization holds your record. Each one holds a fragment, written in its own language, for its own purposes, and stops maintaining it the day you leave. Whatever continuity exists across those fragments exists only if you kept it.

And the search itself is long. Statistics Canada's August 2026 Labour Force Survey(opens in new tab) put unemployment at 6.4% and found that 24% of unemployed Canadians had been looking for 27 weeks or more. A record you build in one weekend has to hold up for six months of postings, each one asking a slightly different question of the same twenty years. It has to be rebuilt again for each one, or it has to have been kept properly in the first place.

Careful doesn't fix this. A careful reaction is still a reaction. It just takes longer.

The slow reaction nobody notices

There's a second reaction, and it's the one I ran for most of my career without knowing it was one.

Every version of my résumé was tailored to a job I never landed. That's how it goes when you're looking, whether it's a full-time role or a contract: a posting comes up, you tailor the document to it, you send it. Then the tailored version becomes the master, and the next posting gets tailored from that. Every edit is reasonable on its own. There's no baseline to compare against, so nothing ever looks wrong. It took me a decade to notice, across hundreds of versions on hard drives spanning ten computers, that the document no longer sounded like me or like the work I'd actually delivered. I hadn't lost the details in one rebuild. I'd edited them out, a line at a time, toward jobs that never happened.

"I hadn't lost the details in one rebuild. I'd edited them out, a line at a time, toward jobs that never happened."

The weekend rebuild and the decade of drift look like opposites. One is fast and panicked, the other slow and diligent. They're the same mistake: the résumé was being used as the record. A résumé is an output. It's a view of the record, cut for one reader and one posting, and it should be cut differently every time. When the output is also the source, every cut is permanent, and the thing you were cutting from gets smaller with each one.

So a regime is not "update your résumé more often." That's the drift with a faster clock. A career maintenance regime is a record that sits underneath the résumé and never gets tailored to anyone, and it has three properties that a reaction, fast or slow, never does.

It runs on a trigger, not a mood. Something changes, and the change gets written down that week. The scope shift, the number that came in, the call you made and why. Not because you feel like updating anything, and not because a posting appeared, but because the trigger fired. Ten minutes when something changes(opens in new tab), as the first post in this series put it. The reaction waits for an event that concerns you. The regime fires on events that concern the work.

It keeps the reasoning, not just the outcome. The outcome survives on its own. It's in the performance review, the project close-out, the two-line bullet you'll write anyway. The reasoning is what evaporates, and it's what makes the outcome yours rather than the team's. The regime captures the constraint, the option you rejected and the reason the timing mattered while they're still specific, which is the week they happened and not the September three years later.

It belongs to you. Your employer keeps a record of you, and it's a worse one than you'd think. A study of job performance ratings published in the Journal of Applied Psychology(opens in new tab) found that roughly 62% of the variance in ratings was down to the rater, not the person being rated. The organization's record of your work is mostly a record of who happened to be reviewing it. The regime is the only record where the person who did the work is also the person keeping the account, and where the account survives the employer.

What the series has been saying

Six posts in, here's the shape of it.

The habit is small: ten or fifteen minutes when something changes, not a quarterly project. The performance review doesn't capture the decisions(opens in new tab), only the results, so if you want the decisions kept you keep them. Long tenure isn't the risk(opens in new tab); undocumented tenure is, and growth you can't prove is growth you can't use. Rebuilding costs you in weeks what maintaining would have cost in minutes(opens in new tab), and the part that doesn't come back at any price is the decision context. And when you finally submit the document, the first reader is a machine(opens in new tab) that indexes what's written and infers nothing.

Each of those is a property of a regime. Together they're the argument for one. Kept on a trigger, the record is current. Kept with the reasoning, it's specific. Kept by you, it survives the reorg, the recruiter and the next employer. Cut into a résumé for one posting, it produces a document that sounds like you and matches what you delivered, because it was cut from something that never drifted. And the following week, when a different posting asks a different question, you cut again from the same record instead of from last week's answer.

That's the version of "stay ready(opens in new tab)" I mean. Not a posture, and not a faster reaction. A record that was there before the trigger.

"Not a posture, and not a faster reaction. A record that was there before the trigger."

What a regime looks like when it's software

I built Tenure because I kept running into this problem from the wrong side of it. The premise is simple: your career is an asset, and like any asset, it needs to be maintained. Not just dusted off when you need to sell it. Tenure is the platform built around that idea. It's live at owntenure.ca. Free trial, no credit card required.

The Modules map to the three properties. Tool Filing Assistant is the trigger: Capture is three fields, what you delivered, the number if there is one, and the part only you'd know, filed against the engagement the week it happened. BattleCard holds the reasoning, the constraint and the trade-off and the call, attached to the work and still legible years later. Gap Analyzer runs the currency check against the roles the market is posting now, so you know your own gaps before a recruiter does. Resume Factory cuts the résumé from that record for one posting, and cuts it again from the same record for the next one, which is the whole difference between an output and a master. And every Assembly Session closes with a Session Audit Certificate, a dated audit trail showing the output is grounded in your real work and your own reasoning.

None of that promises you the call. What it promises is that when the call comes, the record is already there and it's yours.

Regime, not reaction

The reorg will happen on its schedule, not yours. The recruiter will call on a Friday. September will come around again next year. None of those is the problem. The problem is meeting any of them with a record that has to be built on the spot, or with a résumé that has spent a decade quietly becoming someone else's.

Keep the record on a trigger. Keep the reasoning with the outcome. Keep it yourself. Then cut the résumé from it as often as you like, because the thing you're cutting from isn't going anywhere.

Maintained, not reconstructed. Own your tenure.