A teaching assistant advert on a government jobs board reads "£95 to £115". Most people scanning it will read that as weekly, or assume it is a typo for something annual. It is a day rate. Work 190 school days and it is real money. Work the three days a week the agency actually has, and it is not.
We re-read 541 UK teaching assistant adverts from their original pages. Not the job titles or the search results, the actual advert as published, including the structured fields for salary, hours and contract type that sit above the description.
The thing that jumps out is not what schools ask for. We covered that in our earlier read of 500 teaching assistant adverts, where safeguarding appeared in 79% and SEND in 32%. The thing that jumps out this time is who is advertising, and on what terms.
Two in three advertised posts are not permanent
Of the 500 adverts carrying a contract type, 39.4% were temporary and 25.0% were contract. Permanent posts made up 35.6%. So 64.4% of advertised teaching assistant vacancies on this board were not permanent.
That is not a claim about how schools employ teaching assistants. Plenty of permanent TA posts are filled internally, advertised on a single trust's own site, or never advertised at all. It is a claim about what a person looking for TA work actually sees when they look.
And the pay is quoted by the day
436 of the 541 adverts carried a salary field. Strip out the ones saying "competitive" or "negotiable" and 355 state a real figure. Of those:
- 77.5% quote a day rate. Typically £85 to £120.
- About one in ten quote an annual salary.
- The rest are hourly, or weekly.
A day rate is not a salary with extra steps. It is paid for days worked. No school holidays, no sick pay in most cases, and no guarantee of five days next week. An advertised £110 a day is not comparable to an advertised £22,000 a year, and nothing in the advert helps you make that comparison.
We checked this held up before writing it down. Counting each employer once rather than each advert, 8 of the 12 employers with three or more adverts mostly quote day rates. Removing the single largest advertiser entirely still leaves 71.3%. The pattern is not one agency dragging the average.
43 employers, 541 adverts
This is the number that explains the other two. The whole set comes from 43 distinct advertisers. One supply agency accounts for 140 adverts on its own, more than a quarter of everything we read.
The visible teaching assistant market is mostly supply agencies posting the same kind of role across many schools. That is why the pay is daily, why the contracts are temporary, and why the adverts read the way they do. It is a recruitment channel, not a census of the profession.
It also means a school advertising a permanent, salaried TA post is competing for attention in a feed dominated by day-rate supply work, and the two look almost identical at a glance.
The thing almost none of them tell you
Only 16.3% of the 541 adverts say "term-time only" anywhere. 20.3% mention term-time at all.
For a role that is overwhelmingly term-time in practice, that is the single most consequential fact about the job, and four adverts in five leave it out. Combine that with a day rate and you have an advert that gives a number without giving any way to work out what a year of it pays.
To be fair to the advertisers, the hours field is usually filled in: 92.4% state full time or part time, and 77.4% of those say full time. But "full time" on a supply contract means full time on the days there is work.
What this means if you are applying
Read the salary line for its period before its number. If it says a day rate, you are almost certainly looking at agency supply work, and the questions to ask are how many days a week are guaranteed, whether the placement is one school or several, and what happens in the holidays.
If you want a salaried school post, the filter is the contract type, not the job title. Permanent posts are about a third of what is advertised, and they are the ones quoting an annual figure.
And when you apply, the advert's own words matter more than the polish of your CV. Our earlier analysis found safeguarding in 79% of TA adverts and SEND in 32%. If you have that experience and your CV does not say so in the advert's language, a filter will not find it.
Method
541 UK teaching assistant adverts saved from DWP Find a Job, re-parsed from the stored HTML rather than from an earlier extraction. That distinction matters: our first pass had stored the salary as "£95 to £115" and dropped the "per day" that follows it on the page, which would have produced a confident and wrong article about advertisers hiding the pay period. They do not hide it. We lost it.
Percentages are given against the number of adverts that carry the relevant field, and that denominator is stated each time. Pay period is read from the advert's own wording. The employer-level and leave-the-biggest-agency-out checks are reported above because a corpus of 43 advertisers is small enough for one of them to move a headline.
Limits. One board, one snapshot. Find a Job carries a particular slice of the market, heavy on agencies, and schools advertising directly on their own sites or through a trust are not in here. Nothing above should be read as the share of teaching assistant jobs in England that are temporary. It is the share of advertised ones we could see.
The wider corpus behind this work is described in ten findings from 15,291 job adverts, and the pay-shape question across 33 occupations is in how wide UK advertised salary ranges really are.