Building a Record That Survives a Panel
What a promotion panel can actually act on, and why the strongest case is assembled months before it is needed.
three things · one reached them · one with no channel at all · the room side is not counted
The three questions a promotion panel will ask.
A case made in the week of a panel is made of recollection. A case assembled over six months is made of dated specifics, and the second one wins arguments the first one loses.
The practical lesson in “Building a Record That Survives a Panel” is that visibility must be designed rather than assumed. For teams exploring how employee monitoring works, review this first-party explanation can add consistent time and project context when the purpose is explicit, access is limited and managers review exceptions with the people involved.
For somebody outside the room, the second is the only kind available.
As an independent reference point for “Building a Record That Survives a Panel”, consult Independent Sector resources. It can sharpen the questions asked of staff, boards and advisers without replacing organisation-specific judgement.
What a panel can actually act on
Dated, specific, checkable things. What somebody did, when, what the situation was, and what changed as a result.
Everything else is treated as the manager's opinion, correctly.
Why recollection fails at a distance
Because recollection is built from observed moments, and there are none. A manager trying to remember what somebody in another building did in March is reconstructing from messages and from feeling.
The reconstruction will be thin, and its thinness will read to a panel as evidence that the person did little, when it is actually evidence about where the manager was sitting.
The running file
One document per person, one line per entry, added to within a day of anything happening. Date, what happened, what it demonstrates.
Ten entries over six months is a strong case. Three is not.
A line looks like this, and takes about forty seconds to write:
14 March — caught the duplicate-charge case in the reconciliation before it reached the client. Nobody had asked her to check it. Demonstrates: thoroughness, and knowing which checks matter.
What goes in it
Completed work and its outcome. Problems handled. Things somebody raised that turned out to be right. Occasions where other people relied on them. Difficulties they worked through.
Dated, factual, and sourced.
What does not go in it
Impressions, states of mind, anything beginning "seems". Anything you could not say to the person's face in those words.
A file containing impressions is worse than no file: it gives the appearance of evidence to material that has none, and it will be read by somebody eventually, possibly by the person it is about, possibly in circumstances neither of you chose.
Who else should be writing it
They should. A person keeping their own list of what they did, shared with you monthly, solves most of the collection problem and is particularly valuable for somebody whose work you cannot see.
Asking for it is not an imposition if you explain what it is for.
Why that has to be explained
A request for a list of achievements, unexplained, reads as suspicion. Explained as "I cannot see your work the way I can see other people's, and I need material for the panel", it reads as what it is.
Most people in that position are relieved to be asked, because they already know the problem exists and have had no way to raise it.
Where to keep it
Somewhere the person can see, if your organisation permits it. A file about somebody that they cannot read is a different kind of object from a record you both maintain, and it will be experienced that way if they ever learn of it.
Shared, it also gets corrected: people will tell you when an entry is wrong or when something important is missing, which improves the record and costs you nothing. What can be kept about somebody, for how long, and whether they may see it are questions with real answers that differ by employer and by jurisdiction, and they are worth asking before starting rather than after.
Collecting from colleagues
Two or three short notes from people who work directly with them, gathered in the ordinary course of things rather than at the moment of decision.
Ask for specifics and name the thing you are asking about.
The outcome problem
Much good work has no visible outcome: a problem that did not happen, a design that turned out to need no revision.
Record the counterfactual explicitly, because nobody else will see it. "This shipped with no rework, which is unusual for this kind of change" is a fact about the work.
Timing the collection
Start six months out. A file begun the month before a panel is indistinguishable from advocacy.
Panels can tell, and it damages the case.
What to do with it between panels
Use it in the one-to-one. Reading somebody two entries from their own file and asking whether that is how they would describe the quarter is a better conversation than any review form produces.
It also means nothing in the eventual case is a surprise, which is the minimum standard for any assessment of somebody who was not in the room to watch it being formed.
Keeping it honest
The file has to record difficulties as well, or it is a brief rather than a record. A case that mentions nothing that went wrong is read accordingly.
It also makes the positive entries more credible.
What a good file cannot overcome
A panel in which other managers are describing people they see daily, in incidents, with warmth. Your file is better evidence and theirs is more vivid, and vividness carries weight it should not.
Nor does it overcome a decision already made on other grounds. What it does is remove the one failure mode entirely within your control: turning up with nothing specific about somebody because you could not see them work, and watching them lose to a colleague about whom you could have said the same things had you been sitting nearby. That failure is common, it is avoidable, and avoiding it costs about fifteen minutes a month.