Running a waiting list that people believe in
Waiting lists are unusual among administrative tasks in that the hard part is not the record-keeping. It is the fairness, and the fact that fairness has to be visible to be worth anything. A queue people cannot check is a queue people assume is rigged. This piece covers how to order a list defensibly, why a position number tells someone almost nothing, how to work out a wait people can actually plan around, and what to do about the awkward cases every real list produces.
Write the rule down first
Every waiting list runs on a policy, whether or not anybody has written it. The policy might be pure order of arrival, or it might be arrival within priority bands, or it might be a discretionary judgement made afresh each time. All three are defensible. Only the first two are checkable.
The value of writing it down is not bureaucratic. It is that a person who can reconstruct their own position from rules they can see has no reason to suspect anything, and a person who cannot has every reason. Most complaints about waiting lists are really complaints about opacity.
The simplest defensible rule is two-part: sort by priority band, then by the moment of joining. Two visible columns, no hidden weighting, and an order anyone can verify with a pencil. Anything more sophisticated should have to earn its place against that.
Priority bands need to be few and meaningful
Bands exist so that genuine urgency can move ahead of ordinary need. They stop working the moment they become a general expression of how strongly somebody feels about their case.
Three or four bands is usually right. More than that and the boundaries stop being distinguishable, so the assignment becomes arbitrary and the whole structure loses its authority. Fewer, and there is nowhere to put a real emergency.
Whatever the bands are, they need written criteria and somebody willing to say no. A list where most entries end up in the top band has, in effect, no bands at all, and has traded a checkable rule for an unchecked one.
A position is not a waiting time
Telling somebody they are twelfth is telling them almost nothing. Twelfth on a list that clears three people a week is a month. Twelfth on a list that clears three a year is four years. The number is identical and the meaning is completely different.
Waiting time is position divided by the rate at which places actually come free, and that rate is a property of the service rather than the list. It has to be measured, not assumed, and it is the single most useful figure an administrator can have.
This matters practically as well as intellectually. People make decisions on the basis of what they are told: whether to look elsewhere, whether to make other arrangements, whether to stay on the list at all. A position number given without a rate invites them to guess, and they will guess optimistically.
Measuring how fast a queue moves
The rate comes from your own records: how many offers were made, across how long a period. If eight offers went out over sixteen weeks, places come free at about one every two weeks, and that figure carries far more information than the length of the list.
There is a small trap in the arithmetic. Eight offers spread across a period occupy seven intervals, not eight, and dividing by the wrong one understates the rate. It is a minor error on a long history and a large one on a short one.
Short histories are the real problem. Two or three offers say nothing reliable about the fourth, and a rate calculated from them will be confidently wrong. It is better to say that there is not yet enough information than to publish a number that will be quoted back at you.
Recording offers is the part that gets skipped
Everything above depends on knowing when offers were made, and that is exactly the record most lists do not keep. People note who joined and who eventually came in, and the moment in between disappears.
It is worth capturing because it is the only evidence of how the service actually behaves, and because it separates two very different situations: a list that is not moving, and a list that is moving but where most offers are declined. Those need opposite responses.
The habit is easier to keep if it happens at the moment of contact rather than afterwards. Marking somebody as offered when you ring them costs a second; reconstructing it from an inbox three months later costs an afternoon.
Declines, no-shows and holds
Not everyone offered a place takes it. Circumstances change, people find alternatives, and some join several lists at once precisely because none of them can promise anything. A healthy list has a steady trickle of declines and that is not a failure.
What matters is deciding in advance what a decline means. Does the person leave the list, go to the back, or keep their place? All three are used in practice and all three are defensible; what is not defensible is deciding case by case after the fact, because that is where the appearance of favouritism comes from.
A hold status is worth having as a separate thing. Somebody who cannot take a place this month but wants to stay in the queue is not the same as somebody who has declined, and treating them as one loses information you will want later.
Keeping the list honest over time
Waiting lists rot. People move, change telephone numbers, solve their problem elsewhere, or simply forget they applied. A list that has never been cleaned overstates demand, sometimes by a lot, and overstated demand leads to bad decisions about capacity.
A periodic check — an annual message asking people to confirm they still want to be on the list — is the standard remedy. It has to be done carefully, since a check that removes anyone who does not reply will remove people who never received the message, and those tend to be the ones least able to chase it.
The conservative version is to ask, wait, ask again through a different channel, and only then move somebody to a dormant state rather than deleting them. Cleaning a list is worth doing; losing people from it is not.
What to tell people, and how often
Silence is the thing people find hardest. A list that communicates once a year is experienced as a list that has forgotten you, even when the position has not moved and there was genuinely nothing to say.
A short, regular update that includes both the position and the current rate is more reassuring than an occasional detailed one, and it has the useful side effect of keeping contact details fresh. Say what has changed, say what has not, and be clear that an estimate is an estimate.
It is also worth being honest when the answer is bad. People can plan around a long wait if they are told about it. They cannot plan around a vague reassurance that turns into a long wait, and the resentment produced by the second is far greater than the disappointment produced by the first.
This is personal data, and it behaves like it
A waiting list is a file of names, contact details, dates and often notes about circumstances — health, housing, family situation. That combination is exactly what data protection rules exist for, and it is a poor candidate for being typed into whatever free tool came up first in a search.
The practical questions are the ordinary ones: who can see it, where does it live, how long is it kept after somebody leaves the queue, and what happens to it if the person who maintains it is away. Most small organisations have never answered any of them.
Keeping the data on a machine you control, and taking your own backups rather than relying on a service, answers several of those questions at once. It also removes the possibility that a list which was never meant to leave one office ends up somewhere it cannot be recalled from.
The one duty that survives every arrangement is retention. Records kept for ever because nobody decided otherwise are a liability rather than an asset, and the notes attached to a queue are often the most sensitive part of it. Pick a period after somebody leaves the queue, write it down, and delete on schedule.