The longer a UK job advert gets, the less of it is about the job

The median UK job advert is 408 words. Some run to two hundred, some to nearly a thousand, and the assumption behind that spread is that a longer advert tells you more. So we opened up 505 adverts, split each into its own sections, and counted where the words actually go.

They do not go where you would expect. As adverts get longer, the section describing what the employer wants from you barely grows. The section listing the perks grows more than five times over.

Where the extra words go

Comparing the shortest third of adverts with the longest third, the advert itself grows roughly two and a half times, from a median of 259 words to 665. The growth inside is wildly uneven:

Benefits grows 5.2 times. Duties 2.5 times. The opening pitch doubles. Requirements, the part telling you what they actually want, grows least of all at 1.8 times.

In the longest adverts, the perks are the biggest thing on the page

Read as a share of the advert rather than as raw words, the picture inverts completely.

In the shortest adverts, requirements is the biggest section and benefits the smallest. In the longest adverts that flips: benefits becomes the single biggest thing on the page, bigger than the job description itself, while requirements shrinks to under a fifth.

A long advert, on this evidence, is not a more informative advert. It is a more promotional one.

Why that matters when you are applying

The requirements section is the part you use to answer the only question that matters when you first open an advert: is this me, or am I about to waste an evening? When that section stays thin while the perks around it swell, a longer advert has not made your decision any easier.

So a useful habit: when you open a long advert, scroll straight to the requirements and read those first. If it is four vague bullet points wrapped in three hundred words of culture and cycle-to-work, you are looking at an advert that has not decided who it wants. Worth knowing before you spend an hour on the application, and often a sign the employer will consider a wider range of backgrounds than you would assume.

We also tested the rule everyone repeats

Recruitment advice says keep adverts short or candidates will not read them. Since Reed publishes how many applications each advert has received, we could check rather than guess.

There is no relationship. The shortest quarter of adverts drew 0.62 applications a day and the longest quarter 0.59. Within individual occupations it was a coin flip: shorter adverts pulled more applications in seven roles and fewer in five. Advert length simply does not predict how many people apply. Occupation does, and enormously, as our competition gap research shows.

So this is not an argument that long adverts are bad. It is a finding about what those extra words are spent on.

Length by sector, and why it varies

The variation between occupations does not follow how complex or senior a role is. It follows sector habit. Retail assistant adverts run more than twice the length of cleaner adverts for work of broadly similar scope.

Nobody decided that retail needs twice as many words as cleaning. Each sector inherited a house style and kept writing to it, the same pattern we found when we measured where "fast-paced" actually appears: it clusters where it is habitual, not where the work is genuinely relentless.

What to do with this

Whatever the length, the advert is your brief. Those 408 median words contain roughly eight to twelve things the employer cares about, and most of them sit in the requirements section however small it has become. Your job is to evidence those things quickly and visibly.

That is also the honest answer to the question people ask far more often, which is how long their own CV should be. Cut by relevance to the advert rather than to hit a page count. We go through that properly in our guide on how long a UK CV should be.

Method

We pulled 1,807 full UK job adverts across 12 occupations from the Reed jobs API on 17 July 2026, then restricted the analysis to the 1,243 posted within the previous 30 days, so long-running evergreen listings could not distort application counts. Word counts come from the full advert text. Section analysis was run on the 505 adverts (41%) that could be reliably segmented using the employers' own headings; adverts written without clear section breaks were excluded rather than guessed at, so the section figures describe adverts that use headings, not all adverts. Section comparisons use means, which handle adverts missing a section more honestly than medians do. The length-versus-applications test divides each advert's published application count by its days live, and was run within each occupation as well as pooled, because occupations differ so widely in application volume that a pooled-only result would be confounded. This compares different adverts at one moment in time; it does not measure adverts changing over time. All figures are aggregates and no advert text is reproduced.