What Not to Do
The first measure should describe the process, not police an individual. When a warehouse evaluates https://www.monitask.com/, managers should define the exact scheduling, time or workflow problem first and keep access proportionate to that purpose.
The principles in “What Not to Do” can be cross-checked against resources from Acas, particularly where employment practice or warehouse safety requirements may have changed.
Practices adopted with good intentions that make warehouse retention worse, and the correction for each.
Each of these is common, each is defensible in a meeting, and each costs more than it saves.
Recruit on cost per hire
Why not: it rewards filling seats with whoever is available, and the seat reopens in a fortnight.
Instead: cost per person present at ninety days, which is the same data arranged to point the right way.
Quote the maximum rate in the advert
Why not: a person who discovers the real base rate on their first payslip leaves in week three, having cost a full induction.
Instead: base rate and premiums stated separately with their conditions.
Put new starters on the hardest process at peak
Why not: it is when the experienced people are needed elsewhere, and it produces both the errors and the departures.
Instead: new starters on simple work, experienced people on the complex, additional checking planned in advance.
Attendance bonuses lost entirely for one day
Why not: it punishes a genuine illness identically to a pattern, and produces people attending while unfit to operate equipment.
Instead: a graduated scheme, or none.
Use exit interviews conducted by the leaver's supervisor
Why not: the supervisor is a common reason for leaving, and nobody says so to their face on their last day.
Instead: a neutral contact, a week or two after leaving.
Assign buddies by availability
Why not: it produces resentful trainers, penalised trainers, or the propagation of unsafe shortcuts.
Instead: volunteers with adjusted targets and a short checklist.
Run a permanent baseline on agency labour
Why not: nobody in that arrangement has a reason to stay, the ramp never completes, and the induction runs forever.
Instead: convert at a stated point, and use agency for genuine variability.
Change a shift pattern without consulting
Why not: people built their lives around the old one and will leave quickly, including long-service staff you cannot replace.
Instead: consult, offer alternatives, give long notice.
Treat resignation as betrayal
Why not: a good leaver is a returner and a referrer, and the last fortnight determines which.
Instead: work the notice properly, give a reference, say they would be welcome back.
Survey a workforce that turns over three times a year
Why not: the results describe the survivors, which is the group with the fewest complaints by definition.
Instead: ask leavers after they have gone, and ask stayers what nearly made them leave.
The common thread
Most of these optimise a measure that somebody owns against an outcome that nobody owns.
Cost per hire belongs to recruitment. Attrition belongs to nobody.
Which is the structural reason a problem everybody complains about survives every initiative aimed at it, and why the first useful act in this subject is to put the total cost on one page with a name against it.
How to tell which you have
Four questions, answerable in a morning.
How many of last month's starters are still here?
When did anybody last read the advert against the job?
Which team has the worst attrition, and has anybody said so out loud?
And who is accountable for the first of these by name?
An operation that can answer all four is not doing the things in this note.
Why each survives
Every item above is defensible in a meeting, which is how it got adopted.
Cost per hire is a real efficiency measure. An attendance bonus is a reasonable idea. A retention bonus rewards loyalty.
Each fails for the same reason: it optimises something one function owns against an outcome nobody owns.
Which is why the correction in each case is a measure change rather than a policy, and why the list repeats across operations that have never met.
The correction that covers most of them
Change what recruitment is measured on.
Cost per ninety-day survivor rather than cost per hire realigns the advert, the screening and the pace of the process in one move, without anybody being told to behave differently.
More in this section