The Bradford Factor is a scoring convention that multiplies the number of separate absence spells, squared, by the total number of days absent in a rolling period. It exists because a flat absence percentage treats one twenty-day absence and twenty one-day absences identically, and many employers believe those two patterns disrupt work very differently. Squaring the spell count encodes that belief numerically. It is a belief, not a measurement, and this page is written to keep that distinction visible.
Arb Digital built this calculator because the arithmetic is easy to get wrong in exactly the way that matters: people count days rather than spells, or count each day of a continuous absence as a separate spell, and the score is enormously sensitive to that error. The tool takes a list of spells so it can do the counting for you, and it publishes no trigger value, no threshold and no banding, because those are decisions an employer makes with its own advice. Alongside it, our absence rate calculator covers the unweighted view and our employee turnover rate calculator covers people leaving rather than people away.
What This Bradford Factor Calculator Does
It reads a list of absence spells, counts them, sums their lengths and applies the standard relation S² × D. It reports the score, the two components that produced it, the average spell length, and the score the same total absence would have produced if it had been taken as a single continuous spell. That last figure is the honest way to see what the squaring is actually doing, because it isolates the frequency penalty from the volume of time lost.
It also compares the standard exponent against an unsquared and a cubed variant — not so you can pick the number you wanted, but so anyone reading a score understands how much of it is a modelling choice. The bars show the same total days rearranged into one, two, three and four spells, which makes the non-linearity obvious.
It produces no verdict. There is no pass, no fail, no red or amber, and no comparison to any threshold except one you type in yourself.
How to Use It
- List the length of each separate absence spell in the rolling period, in days, separated by commas. One continuous absence is one entry, whatever its length.
- Check the spell count and total that appear in the two boxes underneath. If your records only hold totals, switch the mode and type them directly.
- Set the rolling window your policy uses. Fifty-two weeks is the most common, but the window is part of the convention and changes the score by changing what is in scope.
- Enter your organisation's own trigger value if it has one, purely so the tool can show the distance to it. Leave it at zero if there is no trigger.
- Read the one-spell comparison alongside the score. If the two are far apart, the score is being driven by frequency rather than by time lost, and that is what any conversation should be about.
The Formula
The standard relation is:
B = S² × D
where S is the number of separate spells of absence in the rolling period and D is the total number of days absent across those spells. Both counts run over the same window, normally a rolling fifty-two weeks measured backwards from today rather than a fixed calendar or holiday year.
A worked example makes the weighting visible. Four spells lasting one, one, two and three days give S = 4 and D = 7, so B = 16 × 7 = 112. The same seven days taken as one continuous absence give S = 1 and B = 7. Identical time lost, a sixteen-fold difference in score. Seven separate one-day absences would give 49 × 7 = 343, and ten one-day absences give 100 × 10 = 1,000.
Because D grows at least as fast as S when spells are short, a run of single days makes the score grow roughly with the cube of the number of absences. Scores escalate far more sharply than most people expect, and a number that sounds alarming may represent very little lost time.
It Is a Convention, Not a Standard
There is no standards body that defines the Bradford Factor, no regulator that endorses it and no legal status attached to any particular score. It is a formula that circulated through occupational health and HR practice and was adopted because it is simple to compute and easy to explain. Employers choose whether to use it at all, choose the rolling window, choose the trigger points, choose which absence types are counted, and choose what happens when a trigger is reached. Every one of those choices changes the outcome as much as the arithmetic does.
That matters when a score is quoted as an objective fact. It is an objective computation applied to a subjective set of choices, and two employers can look at the same attendance record and legitimately produce scores an order of magnitude apart. Acas publishes guidance on what an absence policy should contain and how it should be applied, in its material on creating absence policies.
What Counts as a Spell — the Detail That Decides the Score
Because S is squared, spell counting dominates everything. Three questions decide it, and a policy that does not answer them explicitly will produce inconsistent scores.
First, are non-working days between absent days part of the same spell? If someone is off Friday and the following Monday, is that one spell or two? Most policies treat a continuous period of incapacity as one spell regardless of intervening non-working days, but this has to be written down.
Second, does a single day back at work break a spell? Someone who returns, cannot manage it and goes home again has had either one spell or two, and the difference between S = 1 and S = 2 quadruples the score.
Third, which absence categories are in scope at all? Statutory family leave, jury service, authorised study leave and absence arising from a disability are commonly excluded, and treating them as spells is where policies get into difficulty. Building the spell list from clean records first, using tools such as our date difference calculator for spell lengths and our work hours calculator for the underlying roster, is more reliable than reconstructing it from memory.
Disability, Pregnancy and the Limits of a Score
Absence scoring interacts directly with employment law, and the interaction is not optional. In the UK, the Equality Act 2010 defines a disabled person as someone with a physical or mental impairment that has a substantial and long-term adverse effect on their ability to carry out normal day-to-day activities; the official definition of disability under the Equality Act 2010 sets out the test, including conditions that qualify from diagnosis. Where a person is disabled, employers have duties that can extend to adjusting how absence is recorded, scored and triggered — discounting disability-related absence is a commonly considered adjustment.
Pregnancy-related absence is treated separately again in many jurisdictions, and counting it toward an absence trigger can amount to unlawful discrimination. Equivalent protections exist under different names in other countries, with different tests and different scope. A calculator cannot apply any of this. It can only tell you that the categories exist and that the spell list you feed it should already reflect your policy and your legal advice.
The practical position is straightforward. The score is an input to a conversation, not a decision. Where absence data touches a person's employment, the relevant considerations — occupational health advice, whether an underlying condition is present, whether adjustments would help, the employee's own account — sit entirely outside the arithmetic. Your HR team and an employment-law adviser should be involved before a score is acted on.
Where the Score Genuinely Helps
Used as a signal rather than a sentence, frequency weighting does surface something real. Short, frequent, unplanned absence cannot be planned around, lands on colleagues at no notice, and in shift and customer-facing work carries an immediate operational cost. A metric that separates it from planned or long-term absence tells managers something a flat percentage hides.
It is also useful in aggregate. Scoring a whole team reveals whether a unit has a broad pattern of short absences — usually a scheduling, workload or management signal — or a few long-term cases, which is a welfare and cover question. Reading it beside the raw rate from the absence rate calculator and the length-of-service picture from the employee tenure calculator diagnoses far better than any single figure. Statistical definitions of sickness absence, including how a national rate is constructed, are published by the Office for National Statistics in its sickness absence in the UK labour market release.
The Known Criticisms
The Bradford Factor is widely criticised, and anyone using it should know why. It penalises people with fluctuating conditions such as migraine, asthma, inflammatory bowel disease or mental-health conditions, which by their nature produce short, frequent absences. It creates an incentive to attend while unwell or to extend a short absence into a longer one to avoid a second spell, both of which are worse for the employer than the behaviour being discouraged. It carries no information about cause, and it treats a person with a single bad winter identically to a persistent pattern.
None of that makes the arithmetic wrong. It makes the score a poor basis for consequences applied mechanically, which is precisely the use it most often gets put to.
Arb Digital builds free tools like this one because useful pages earn attention. If you want tools, calculators or content built for your own audience, we can help.
Browse All Free Tools Talk to Arb DigitalCommon Mistakes to Avoid
- Counting days as spells — a five-day continuous absence is one spell. Entering it as five multiplies the score twenty-five-fold and is the single most common error.
- Using a fixed year instead of a rolling window — a fixed year resets everyone on the same date and lets a pattern straddling the boundary disappear, which is not what the metric is for.
- Scoring protected absence — pregnancy-related and disability-related absence are treated differently under employment law in many jurisdictions, and including them creates both a distorted score and legal exposure.
- Reading the score as an amount of lost time — it is not one. A score of 1,000 can be ten days, and a score of 7 can also be seven days.
- Applying a trigger automatically — the number identifies a record worth looking at. What is behind the record is the thing that matters, and only a conversation reveals it.
Related Free Tools From Arb Digital
Measure unweighted time lost with the absence rate calculator, track people leaving with the employee turnover rate calculator, and look at length of service with the employee tenure calculator. Price the roles involved with the employee cost calculator, value untaken leave with the PTO payout calculator, measure spell lengths with the date difference calculator, and browse the full free online tools hub for the rest of the library.
Frequently Asked Questions
Score equals the number of separate absence spells squared, multiplied by the total number of days absent, both counted over the same rolling period. Written out: B = S² × D. Four spells totalling seven days give 4² × 7, which is 112.
No. It is a scoring convention that individual employers choose to adopt. No standards body defines it, no regulator endorses it, and it has no legal status. The window, the trigger points, the absence categories in scope and the consequences are all set locally.
This page will not answer that, because thresholds are set by each employer's own policy and legal advice, not by arithmetic. You can type your organisation's own trigger into the tool and it will show the distance to it, but it supplies no value of its own.
That depends on your policy, and it must be written down, because the answer changes the score fourfold. Most policies treat a continuous period of incapacity as one spell even when non-working days fall inside it, but the convention has to be stated rather than assumed.
It is treated differently under employment law in many jurisdictions, and discounting or adjusting it is a commonly considered reasonable adjustment. This calculator cannot make that judgement. Exclude what your policy and your legal advice say should be excluded before you build the spell list.
Because the spell count is squared and short absences push the day count up at the same time. A run of single-day absences makes the score grow roughly with the cube of the number of absences, so scores in the hundreds or thousands can represent a very small amount of lost time.
The score is an input to a conversation, not a decision. It carries no information about cause, no view on whether adjustments would help and no legal analysis. Any process with consequences for an individual needs HR involvement and, where absence may be health- or pregnancy-related, employment-law advice.
To make the modelling choice visible. Comparing S × D, S² × D and S³ × D on the same absence shows how much of any score is produced by the exponent rather than by the person's attendance. The standard convention is the squared one.
This tool is provided for informational and educational use only. It is not employment advice, medical advice or legal advice, it sets no thresholds and it reaches no conclusions about any individual. Absence scoring interacts with disability and pregnancy discrimination law, and the rules differ by jurisdiction. Consult your HR team and a qualified employment-law adviser before a score informs any decision affecting a person's employment.
