Keeping a Bank That Answers
A pool of people who work occasional shifts is the difference between a forty-minute morning and an agency booking. Most banks exist on paper and not on the phone.
A bank of eleven, one Tuesday
- Worked in the last 3 months — 4 of 11Accepted
- Not worked in 6 months — 5 of 11No answer
- Competency lapsed — 2 of 11Declined
- Answered the phone this morning — 2Accepted
- Available — 1Accepted
A bank of eleven produced one available person. The list was last reviewed when it was created.
A bank is a group of people who will work occasional shifts without a guaranteed pattern. Where it works it is the best cover route after the free ones: known people, known competence, a rate below agency, and no induction.
The staffing problem described in “Keeping a Bank That Answers” should be separated from assumptions about individual effort. When an organisation evaluates how to monitor employees without being intrusive for how to monitor employees without being intrusive, it can use time and project records to understand capacity, provided schedules, approved leave, exceptions and the correction route remain part of the same decision.
Where it does not work — which is most places that have one — it is a list of names collected at some point and never maintained, producing a morning that feels like it has options and does not.
For an independent benchmark relevant to “Keeping a Bank That Answers”, consult the CIPD working-time guidance. Use it to test scheduling, working-time limits, attendance records, employee rights and exception handling against the real operation rather than treating a software report as self-explanatory evidence.
What makes a bank real
People who have worked recently. A bank member who has not done a shift in six months is not a bank member; they are a former employee with a number on a sheet. Three months is a workable threshold and people past it should be contacted deliberately rather than at six forty.
Current competence. Certifications expire, systems change, procedures move on. A bank worker returning after eight months may need a morning of updating before they can be useful, which is not available in the hour.
Current access. Logins disabled for inactivity, badges deactivated, keys returned. Discovering this when they arrive is a routine and avoidable failure.
A rate they are happy with, agreed in advance and not negotiated on the phone.
Keeping it warm
Offer shifts before you need them. A bank used only in emergencies decays, because the relationship is only ever a request made under pressure.
Where there is predictable demand — a known busy period, a planned absence, a training day — give it to the bank rather than to overtime. It keeps competence current, keeps access alive, and means the emergency call is to somebody who worked a fortnight ago.
Keep them informed. Bank workers are the first to be left off every communication, which is how they find out the procedure changed by getting it wrong.
Recruiting to it
The best source is people already leaving or reducing: retirees, people going part-time, people moving to study, parents reducing hours. They are competent, known, and frequently willing.
Ask at the exit conversation rather than six months later. The conversion rate from "would you like to stay on the bank" asked on the last day is very much higher than from a cold approach afterwards.
Students and people with another job are the other main source, and both come with hours and rest constraints that must be tracked rather than assumed.
The employment questions
Bank arrangements sit in genuinely contested territory in many jurisdictions. Whether somebody on a long-running bank arrangement is a worker or an employee, what continuity of service they accrue, what entitlements follow, and whether a pattern of regular shifts has created something other than casual work — all of these depend on the facts and on where you are.
Two practical rules hold anywhere. Write down what the arrangement is, and check it against what actually happens once a year, because what actually happens drifts. A bank worker doing three shifts a week for two years is not casual whatever the paperwork says.
The size to aim for
Enough that the expected number available on any morning is at least two. That is a function of bank size and of how often each member is willing to work, both of which you can measure.
A bank of eleven that produces one available person is the common state and it is worse than useless, because it creates the belief that cover exists. Counting the realistic number — recent, current, reachable — is the first thing to do, and in most operations it is a quarter of the headline figure.
The review
Quarterly: who has worked, whose competencies are current, whose access is live, who should be contacted or removed.
Twenty minutes. It is the difference between the list in the example above and a bank that answers.
What the bank is worth
Compare, over a quarter, what bank cover cost against what the same shifts would have cost through agency. The ratio is usually between a third and a half.
That figure is the justification for the maintenance effort, for the induction time, and for offering shifts to the bank when overtime would have been easier. Without it the bank is an administrative overhead that somebody will eventually propose closing, and the proposal will look sensible on the numbers that are visible.