Preparing for Your Annual Review: Why the Night Before Is Too Late
The advice is everywhere and it never changes. Assess yourself honestly, gather your evidence, print the congratulatory emails, build a portfolio of wins, and adopt a consensual, proactive attitude.
None of that is wrong. It just arrives too late, and it rests on an assumption nobody states out loud: that the annual review measures your contribution. That is not what it measures.
What the rating actually measures
There is an old and solid finding on this. In 2000, Scullen, Mount and Goff published an analysis in the Journal of Applied Psychology of two large data sets, 2,350 and then 2,142 managers, each rated by seven people: two bosses, two peers, two subordinates and themselves.
Their conclusion: idiosyncratic rater effects accounted for 62% and 53% of the variance in ratings depending on the data set. More than half. Every rater has their own way of scoring. Some are generous, some are harsh, some use the whole scale and others cluster everything in the middle.
A performance rating therefore tells you more about who is rating than about who is rated. That is not a reason to dismiss the exercise, since the rating feeds your raise and your progression. It is a reason to stop reading it as a verdict on your worth, and to treat it as what it is: a noisy instrument whose inputs you can improve.
”Gather your evidence” arrives too late
The first input is your manager’s memory, and a year is not remembered evenly. The last few weeks weigh more than the first, one striking incident weighs more than six steady months, and anything that left no trace weighs nothing at all.
This is where the standard advice turns against certain profiles. A good part of real work produces no presentable artefact. Fixing a problem before it breaks, unblocking a colleague, rewriting a confused specification to avoid three weeks of pointless development, absorbing a load nobody saw coming. None of that prints.
People whose contribution runs through those channels are not performing less. They are documented less. We cover the same gap on the hiring side in what interviews actually measure, and on the skills side in identifying your hidden talents.
Why asking for more specific feedback is not enough
The obvious objection would be to ask your manager for more concrete feedback. A large-scale experiment suggests it does not resolve that easily.
In 2021 the Behavioural Insights Team ran a randomised controlled trial on the 360 degree reviews of 4,328 senior managers in a large UK public sector organisation, involving 25,627 reviewers. Half saw the usual interface, the other half an interface that explicitly asked for specific examples and advice on how to improve.
The result is instructive because it is disappointing. The intervention increased the specificity of answers on the overall question, changed nothing on the development question, and actually decreased it on the strengths question. A design built specifically to obtain more precise feedback only half worked.
The practical lesson is clear: do not rely on the instrument, or on your manager’s goodwill, to produce the material. Produce it yourself.
Keep a record, one line a week
The method that works is boring and takes five minutes a week. One line, three pieces of information: what changed, for whom, and how you know.
“Refactored the billing module” is an activity. “Billing errors went from twelve a month to two, and the support team confirmed it in June” is an outcome, with a witness and an order of magnitude. The difference is not cosmetic: the second version survives inside someone else’s memory.
Two things matter here. Record the work that has no artefact too, precisely because that is what evaporates. And feed it upward in small doses through the year, two lines in a monthly check-in, rather than dumping twelve months of record the night before the review. A year of achievements revealed all at once reads as advocacy; the same information distributed reads as reporting.
It is also the material you will need if you are aiming for an internal move, which we cover in the internal mobility playbook.
Visible or accessible
The same report contains a finding its own authors call tentative, and it is worth knowing: rating someone on how accessible they are correlates better with overall performance than rating them on how visible they are. The authors go on to recommend that employers avoid questions about work inputs, such as hours worked or location.
If the finding holds, it usefully shifts the effort. Making yourself visible demands a continuous social performance, expensive and unevenly distributed, which we set out in masking and its career cost. Being reachable at the right moment is a different requirement, plainer and more sustainable.
An annual review will remain an imperfect instrument, and preparing for it guarantees no raise. But you are not obliged to let someone else’s memory be the only record of your year.
Prepare your next step. Discover your profile on NextWorkStep.