Back to blog
Compliance8 min read

Is the Bradford Factor Legal? Yes — And That’s the Problem

Desk scene with a Sickness Absence Policy document highlighting the Bradford Factor formula S squared times D, stamped "Legal but risky", beside a score card reading 5 absences times 2 days equals 250 points, a calculator, a laptop showing an HR compliance dashboard, and a clipboard checklist: discount disability-related absence, exclude pregnancy-related absence, human review before any action

The most common absence trigger in UK handbooks is perfectly lawful. It's also the single most reliable way for a small employer to walk into an Equality Act claim without noticing.

Someone asks me a version of this question most months: "Is the Bradford Factor legal?" They usually expect a yes or a no. The honest answer is worse than either: yes, it's legal, and that's exactly why it keeps hurting people. Nothing about the formula warns you where it stops being safe.

What the Bradford Factor Actually Is

The Bradford Factor is a scoring formula: S² × D — the number of separate spells of absence, squared, multiplied by the total days absent. It deliberately punishes frequency over duration:

  • One absence of ten days: 1 × 1 × 10 = 10 points
  • Five absences of two days each: 5 × 5 × 10 = 250 points
  • Ten single days: 10 × 10 × 10 = 1,000 points

Same ten days off. Wildly different scores. The theory — it came out of research at Bradford University's management school in the 1980s — is that short, frequent absence is more disruptive and more likely to be discretionary than one long, genuine illness. Handbooks then bolt trigger bands onto the score: commonly around 50 points for an informal chat, 200 for a formal review, 400+ for disciplinary territory.

No statute mentions the Bradford Factor. No case has ruled the formula itself unlawful. If you want to calculate it for every employee, you can.

Where It Crosses the Line

The trouble is never the arithmetic. It's what the arithmetic gets connected to. There are three well-worn routes from "lawful scoring formula" to "tribunal claim," and most policies I read are exposed on at least one.

1. Disability-Related Absence Drives the Score

Think about what S² rewards: it hammers frequent, short absence. Now think about what disability-related absence often looks like: migraine, asthma, Crohn's, endometriosis, anxiety, chemotherapy cycles. Frequent. Short. Recurring. The Bradford Factor is practically engineered to score disabled employees into the disciplinary band.

Under section 15 of the Equality Act 2010 (discrimination arising from disability), treating someone unfavourably because of something arising from their disability — like absence — is unlawful unless you can objectively justify it. And under sections 20–21, you have a duty to make reasonable adjustments, which tribunals have repeatedly read as including adjusting the trigger itself: discounting disability-related absence from the calculation or raising the threshold for that employee.

The pattern in the case law is consistent: employers lose not because they used a score, but because the score was applied mechanically — the warning or dismissal followed the number, and nobody stopped to ask why the absences were happening. I wrote up a worked example of exactly this failure in why your sickness absence policy is probably illegal — a 2019 handbook, a Bradford trigger, and a £4,200 ACAS settlement.

2. Pregnancy-Related Absence Counts Towards Triggers

Simpler and stricter: pregnancy-related sickness absence is protected under section 18 of the Equality Act, and there is no justification defence. If your policy counts pregnancy-related absence towards any trigger — Bradford or otherwise — it is wrong. Full stop. The policy needs to say these absences are recorded separately and excluded from the score.

3. The Score Acts Without a Human

Plenty of HR systems now calculate Bradford scores automatically and some will generate the warning letter for you. A score that triggers a decision with no human review is where UK GDPR Article 22 on automated decision-making starts to bite: employees have rights to human intervention, an explanation of the logic, and a route to contest the outcome. A manager who reads the flag and decides is fine. A workflow that fires the sanction is not.

Why This Matters More From January 2027

Until now, most Bradford-led warnings landed on employees with under two years' service — people who couldn't claim ordinary unfair dismissal, so sloppy process rarely got tested. That maths changes on 1 January 2027, when the unfair dismissal qualifying period drops to six months under the Employment Rights Act 2025. Far more attendance-related dismissals become tribunal-eligible, and a dismissal that traces back to a mechanical score with no documented human judgment is exactly the kind that fails.

There's a second-order effect worth knowing about too. Since April 2026, statutory sick pay is payable from day one — the three waiting days are gone. Early evidence suggests short absences are now more likely to be taken (and recorded) rather than powered through. If your trigger bands were calibrated in the waiting-days era, expect more employees to hit them without anything about your workforce actually changing. A threshold review is overdue in most handbooks.

What a Defensible Trigger System Looks Like

You don't have to abandon the Bradford Factor. You have to demote it from judge to witness:

  1. Use the score as one input, not the decision. The policy should say a trigger prompts a review, and list what the review considers: patterns, causes, what came up at return-to-work interviews.
  2. Carve out protected absence explicitly. Disability-related and pregnancy-related absence must not count towards triggers. Write the carve-out into the policy — a carve-out applied informally is a carve-out a tribunal never sees evidence of.
  3. Route every trigger through a person. A named role reviews the flag, considers context, and documents the decision — including the decision to do nothing.
  4. Hold return-to-work interviews after every absence. They surface the disability you didn't know about before the score does damage, and they are your evidence that context was considered.
  5. Say all of this in the handbook. The policy employees read must match the process you actually run. A hidden scoring system fails on transparency grounds under UK GDPR before you even reach the Equality Act.

For the full picture of what belongs in the surrounding policy — reporting rules, self-certification, fit notes, SSP mechanics, long-term absence process — see our sickness absence policy guide for UK employers.

The Ten-Minute Check

Open your sickness absence policy and look for three sentences:

  • One that says disability-related absence is discounted or adjusted in any trigger calculation.
  • One that says pregnancy-related absence never counts.
  • One that says a trigger leads to a review by a named human, not an automatic sanction.

If any of the three is missing, your Bradford Factor clause is a liability wearing a metric's clothing. If you'd rather not do the reading yourself, Bounda's free handbook compliance check will flag it — along with every other clause that's drifted out of step with current UK law — with the legislation cited behind each finding.

The formula was never the problem. The autopilot is.

Bradford Factoris the Bradford Factor legalabsence triggerssickness absence policy UKEquality Act 2010reasonable adjustmentsdisability discriminationUK GDPR Article 22

The Bounda Brief

One email a month on UK HR compliance — what's changed, what to action, what to ignore. Unsubscribe in one click.

By subscribing, you agree to the Bounda Privacy Policy. We use Beehiiv to send the newsletter. Unsubscribe in one click. We will never sell your email address.