Every August, I spend one Saturday morning doing something that sounds clinical and has turned out to be the highest-return three hours of my year: I run a performance review on my own career, the way an investor reviews a portfolio. Positions, returns, concentration risk, rebalancing decisions. Coffee, spreadsheet, no self-help language allowed.
I started doing this in 2017, after realizing something embarrassing: I could tell you my checking account balance to the dollar, but I couldn’t tell you whether my earning power — the asset that generates every other asset I’ll ever have — had gone up or down that year. For most working women, career capital is the single largest asset they own; a rough back-of-envelope makes the point. The median full-time working woman earns about $1,098 a week (BLS, first quarter 2026) — roughly $57K a year, which over a 40-year career is well north of $2 million in nominal earnings, before a single raise. Nobody would hold a seven-figure position and never review it. Almost everyone does exactly that with their career.
Here’s the full ritual — the four asset classes, the data pulls, the scoring, and the rebalancing rules — so you can run it on yourself.
A career, like a portfolio, is a set of holdings with different risk profiles and time horizons. I track four:
1. Compensation (the cash yield). What the career pays now: base, bonus, match, priced benefits.
2. Skills (the growth positions). What I can do that the market pays for — and crucially, which of those skills are appreciating versus quietly going obsolete.
3. Relationships (the illiquid holdings). Sponsors, former colleagues, clients, the people who’d take my call. Illiquid, slow to build, and the asset class that has generated every job I’ve gotten since 2014.
4. Evidence (the audit trail). Documented, provable results — the receipts file. Without it, the other three assets are stories; with it, they’re claims that survive diligence.
Investors don’t value holdings at what they paid; they mark to market. Same here.
Compensation: I compute total comp (base + bonus actually paid + employer match + priced benefits), then re-run my market-rate triangulation — fresh BLS OEWS pull for my occupation and metro, a dozen current transparency-state postings, one structured source. The output is a single ratio: my pay ÷ market midpoint. Above 1.0, I’m paid ahead of market; below 0.9, something needs to happen this year. I also log the year’s percentage change against the going rate of wage growth — the Atlanta Fed’s Wage Growth Tracker ran about 3.8% in mid-2026 — because a 2% raise in a 3.8% market is a pay cut wearing a bow.
Skills: I list the five skills my income actually depends on, then ask two cold questions of each: is demand for this rising or falling in the job postings I just read? and when did I last get measurably better at it? Postings are the tell — when I did this in 2023, “campaign analytics” had crept into 9 of the 12 marketing postings I’d pulled for comp data, and my skills row said I hadn’t touched analytics training in three years. That mismatch — visible only because comp research and skills review happened at the same desk on the same morning — dictated the whole next year’s learning plan, and the pay bump that followed it.
Relationships: I count, honestly, the people who would (a) take my call within a week and (b) vouch for specific work. Then the concentration check: how many are all at my current company? A network that’s 80% current-employer is a portfolio that’s 80% one stock — fine until the day it very much isn’t.
Evidence: I reconcile the receipts file — the weekly one-line log of wins and metrics I described in the first-90-days playbook — against the year. Every claim gets a link or it gets cut.
Now the spreadsheet. One row per year; these columns:
| Column | What goes in it | Red flag threshold |
|---|---|---|
| Total comp ($) | Base + bonus paid + match + priced benefits | Growth below wage-growth benchmark |
| Pay ÷ market midpoint | From the mark-to-market | Below 0.90 |
| Skills: rising / total | How many of my 5 core skills show growing demand | Fewer than 3 of 5 |
| New evidence entries | Receipt-file lines added this year | Under 12 — a quiet year or a lazy log |
| Network adds | New people who'd take my call | Zero outside current employer |
| Energy grade (A–F) | Honest gut: does the work still generate energy? | Two consecutive years below C |
That last row isn’t soft. Burnout is a financial event — it degrades performance, then reviews, then raises, then the résumé story, in that order. I grade it like everything else precisely so it can’t hide.
A 2% raise in a 3.8% market is a pay cut wearing a bow.
Reviews without decisions are journaling. The rules I actually follow:
A ritual you have to remember is a ritual you’ll skip, so borrow the mechanics that make retirement contributions work: automation and defaults. Mine looks like this. A recurring calendar block, second Saturday of August, 8–11 a.m., titled “Board meeting” — because that’s what it is; I’m the board. A template copy of last year’s spreadsheet row, pre-filled with blanks, waiting in the same folder. And a standing rule that the review isn’t finished until it produces exactly three written commitments for the next twelve months, each with a date and a first step small enough to do that same week. Not five commitments, not a vision statement — three lines. Last year’s were: refresh the analytics certification by March, two outside-company coffees per quarter, and re-run the comp triangulation before review season instead of after. All three happened, and I’d credit the “first step this week” rule for two of them.
If August feels arbitrary, pick your own trigger — the week after your company’s review cycle closes is a natural choice, because the fresh data (your rating, your raise, the comp letter) flows straight into the spreadsheet while the sting or the shine is still informative. The one timing rule I’d hold firm: do it when nothing is on fire. A career review run mid-crisis becomes a rationalization engine for whatever the crisis is already pushing you toward.
Things this ritual can’t do. It won’t make a bad manager good, and it has a known bias: what’s measurable gets managed, and some of the best career assets — taste, judgment, a reputation for being decent under pressure — resist the spreadsheet. I’ve also watched one friend adopt this system and turn it into an anxiety engine, re-marking her market value monthly like a day trader checking positions. Annual means annual. The whole point of reviewing on a schedule is permission not to think about it the other 364 days.
But the core habit — treating earning power as an asset that gets audited, not a report card that gets received — has been worth more to me than any single negotiation. Negotiations are transactions; the review is the strategy underneath all of them. Three hours, one Saturday, six columns. Your career is very likely a seven-figure holding. Review it like one.