Hiring a cleaner or facilities operative
Four hundred and twenty applications, hours that sit either side of the working day, and a set of CVs that genuinely cannot be told apart.
There is less on these applications than on any others in this list. Cleaning and facilities work draws enormous volume and produces applications with almost nothing in them: previous sites, dates, sometimes a phone number and little else. There is no qualification to compare, no portfolio, and no metric. What actually matters is not on the page and never has been — this person will be in your building unsupervised, often alone, frequently with keys and an alarm code, at hours when nobody else is there. Reliability and trust are the entire hire, and both are invisible. Meanwhile the shift pattern, usually early mornings or evenings, is a hard constraint that removes a large share of applicants and is rarely asked about until far too late.
Hours, access, and what is actually being trusted
Selene establishes the hours and the access level first, because those two facts define this role more completely than any description of the work.
Times, sites, and whether they hold keys
The work itself is largely standard. What varies, and what determines who can do the job, is when it happens, whether there are multiple sites, and how much unsupervised access is involved.
One sentence changes the guide
It also explains why tenure is the useful signal on a role where nothing else is. Someone who cleaned the same site for four years was trusted by somebody for four years, and in a pile of four hundred and twenty applications with no other distinguishing information, a long unsupervised placement is close to the only genuine evidence available.
Exceptional, not just good —
“The one who does the bit nobody inspects. Behind the fridge. Under the desks.”
→ Tenure at a single unsupervised site became the leading scored criterion; breadth of experience stopped counting.
What the guide ends up measuring
Five criteria, and only two of them are scored. On applications this sparse, more criteria would be false precision — the guide would be ranking people on differences the documents do not actually contain.
Two facts, and the hours question does most of the work
Right to work and the specific hours are both plain facts, and on this role the hours question removes more applicants than everything else combined — a six-to-nine evening shift is incompatible with a great many people’s circumstances, and asking it on the application page rather than at interview respects that rather than wasting it.
Scored on tenure and trust, because that is what is there
Selene never receives the candidate’s name or the raw CV. She scores a blind, structured profile, with any stated age, sex, nationality and religion stripped out before scoring runs.
On the sparsest applications in this list she reads for length of unsupervised placement and for references that can speak to reliability, rather than inventing distinctions the documents cannot support.
How the blind scoring works →What stays with you
Selene's half
- Putting the exact hours on the application page, which settles most of four hundred and twenty
- Reading everyone whose hours fit against the same guide, blind
- A shortlist short enough that you can actually meet all of them
- Drafting the declines, so a very large pile still hears back
Yours
- Advertising the role — Selene doesn’t post to job boards or sourcing sites
- Taking up the references, which on a trust hire are the real assessment
- Right-to-work verification before anyone starts; she records what candidates state, she doesn’t verify it
- Every hiring decision, and any override — stored with your reason
Nearby roles
State the exact hours
Not "flexible, evenings" — six until nine. It is the question that resolves four hundred applications.