RosterElf Logo UK
Start trial
UK EMPLOYMENT GUIDE

The Bradford Factor: how it works and where it goes wrong

The Bradford Factor scores absence by how disruptive it is rather than how much of it there is. It is genuinely useful for spotting a pattern across a large team, and genuinely risky when it is wired directly to a disciplinary process.

Key takeaways

  • The Bradford Factor is S² × D — spells of absence squared, multiplied by total days absent, over a rolling period (usually 52 weeks)
  • Squaring the number of spells is the whole point: it weights frequent short absences far more heavily than one long one
  • Ten one-day absences score 1,000. One ten-day absence scores 10. Same days lost, a hundredfold difference in score
  • It is a management tool, not a legal standard — there is nothing in UK law that requires or endorses it
  • Applying it mechanically to disability-related or pregnancy-related absence creates real discrimination risk

The formula

B = S² × D

  • S — the number of separate spells of absence in the period
  • D — the total number of days absent in the period

Normally measured over a rolling 52 weeks, so the score falls away as older absences drop out of the window.

Squaring the spell count is the entire idea. It says that ten separate one-day absences are far more disruptive than one ten-day absence, because each unplanned absence means finding cover at no notice, while a longer absence is planned around after the first day.

The same ten days, three very different scores

Absence pattern Spells (S) Days (D) Calculation Score
One absence of 10 days1101 × 1010
Two absences of 5 days2104 × 10 40
Five absences of 2 days51025 × 10 250
Ten absences of 1 day1010100 × 10 1,000

Four people, ten days lost each, scores ranging from 10 to 1,000. That spread is the tool working as designed — and it is also the reason a score should never be read without knowing what sits behind it.

Trigger points

Employers commonly set escalating thresholds — something like 51 for an informal conversation, 201 for a formal review and 401 for a more serious stage. Those figures are convention, not law. You can set whatever thresholds suit your operation, but you should be able to explain why you chose them and apply them consistently.

Where it creates legal risk

Nothing in UK law requires the Bradford Factor, and nothing prohibits it. The risk is entirely in the application — specifically, in letting a number trigger a consequence without anyone looking at the cause.

Two situations that need care

  • Disability-related absence. A disabled employee is more likely to have frequent short absences, so a mechanical trigger disadvantages them by design. Discounting disability-related absence, or raising the trigger for that individual, may be a reasonable adjustment you are required to make.
  • Pregnancy-related absence. This should be disregarded entirely. Counting it towards an absence trigger is likely to be unlawful, and it is an easy mistake for an automated system to make if nobody has configured an exclusion.

The practical answer is to treat the score as a prompt: a reason to have a conversation about what is happening, not a finding about someone's conduct. Set out the exclusions in the policy, and make sure a human reviews the reason for the absence before anything escalates.

Why more employers are looking at this now

Since 6 April 2026, statutory sick pay is payable from the first day of absence with no earnings threshold. A pattern of single-day absences that previously cost nothing in SSP now costs something every time — see our statutory sick pay guide.

That has made short-frequency absence visible in the payroll as well as on the rota, which is exactly the pattern the Bradford Factor is designed to surface. It is a good reason to measure absence properly. It is not a good reason to automate the consequences.

Getting the data right

The formula needs two numbers per person per year: how many separate spells, and how many days. Both are trivial if absence is recorded consistently against the shift it affected, and essentially unrecoverable if it lives in a mix of text messages, notes and blank cells.

Leave and absence management keeps the record against the employee, and clock-in records show which shifts were actually missed rather than which were planned. If you are still on a spreadsheet, the SICK code in our monthly rota template at least makes the spells countable. Trigger scores are built from health data, so handle them per the UK GDPR rules on employee records — and read probation and performance reviews before a score becomes a decision.

Measure absence properly before you score it

Record sickness against the shift it affected, so spells and days are countable — and the conversation starts from facts.

Start trial Book a demo
4.8 stars by 1,570 users
100+ countries 30,000+ workplaces
Important disclaimer General information only – not legal advice

This guide explains the Bradford Factor as a management tool and is not legal advice. It has no status in UK law, and using it to trigger sanctions carries discrimination risk, particularly in relation to disability and pregnancy. Take advice before adopting or acting on absence triggers. It does not constitute legal, HR, or professional advice and should not be relied on as a substitute for advice specific to your business, workforce, or circumstances.

FAQ

Bradford Factor questions

  • A scoring method that measures the disruption caused by absence rather than the volume of it. The formula is S² × D, where S is the number of separate spells of absence and D is the total number of days absent, normally measured over a rolling 52 weeks.
  • Count the number of separate absence spells in the period, square it, then multiply by the total days absent. Four separate absences totalling six days gives 4² × 6 = 96. The squaring is what makes frequent short absences score so much higher than a single long one.
  • Because the disruption of an absence falls disproportionately at the start of it. An unplanned absence means finding cover at short notice; the second and third days of the same absence are already planned around. Squaring the spell count encodes that view — whether you agree with it is a management judgement, not a mathematical one.
  • There is no standard. Employers commonly set thresholds around 51, 201 and 401, escalating from an informal conversation to a formal process. Those numbers are conventions, not rules — you can set whatever thresholds fit your operation, and you should be able to explain why you chose them.