Reading Fluency Benchmarks UK: What the Numbers Actually Mean for Your School

A quarter of children leave primary school reading at the fluency level most children reach by the end of Key Stage 1.

That is not a guess. It comes from the first large-scale study of reading fluency in England, published by FFT Education Datalab in November 2024. Their Reading Assessment Programme gathered 347,168 assessments from just under 110,000 children across 684 schools between September 2022 and July 2024. Before this, England had decoding data from the phonics screening check and comprehension data from the SATs, and almost nothing on the bridge between them.

What the benchmarks show

Words correct per minute is the standard measure: a child reads an unpractised passage aloud for one minute and you count the words read correctly. It is a blunt instrument, and it captures accuracy and automaticity but not prosody. It also tracks comprehension closely enough to be useful. FFT found a correlation of 0.68 between Year 6 fluency and Key Stage 2 reading comprehension scores.

Here are the median scores. This is what the middle child in each year group reads at each point in the year.

Year group	Autumn	Spring	Summer
Year 1	16	30	43
Year 2	45	60	73
Year 3	68	84	91
Year 4	88	97	101
Year 5	102	108	111
Year 6	108	111	114

Source: FFT Education Datalab, 'Reading fluency in England: interpreting data from the FFT Reading Assessment Programme' (November 2024), Appendix 2.

Growth is front-loaded. A typical child goes from 16 words correct per minute in the autumn of Year 1 to 73 by the summer of Year 2, and then adds far less each year after that. Across Year 5 and Year 6 the median moves by three to six words a term.

Two honest caveats before anyone builds a tracking system on this. The passages get harder as the year groups go up, so the flattening in Key Stage 2 is partly a feature of the assessment rather than a wall the children hit. And the Year 6 summer figure of 114 rests on 5,055 assessments, where every other cell in the table rests on somewhere between fourteen and thirty three thousand. It is the thinnest number in the set and it is doing the heaviest lifting in every headline written about this report, including mine.

These are medians. Half the children score above, half below. The range inside any year group is enormous, and that is where the real story sits.

The gap that should worry you

The headline figures are useful. The quartiles are alarming.

By the end of primary school the lowest quartile of readers have a median of 72 words correct per minute. FFT put it plainly: that is around four years behind the majority of their peers, who reach 73 words correct per minute by the end of Key Stage 1. A quarter of children in England are leaving primary school without the fluency to read for meaning, or for pleasure.

The gap opens early and it does not close on its own. At the end of Key Stage 1, FFT's follow-up analysis identified 3,500 children reading below 45 words correct per minute, and boys, disadvantaged children and summer-born children were all over-represented in that group.

Where the 90 words a minute figure came from

You will still see 90 words correct per minute quoted as the end of Key Stage 1 benchmark. It is worth knowing what it actually says.

It comes from a footnote in the Standards and Testing Agency's key stage 1 teacher assessment frameworks. Roughly 90 words a minute, it says, is a good indicator of the point at which children begin reading fluently enough to attend to meaning, and it adds immediately that some children read more slowly than this and understand perfectly well.

So it was never a target, and it was never a standard. It was an indicator, published with its own caveat attached, and the footnote does not say where the number itself came from.

Its status has shifted since. End of Key Stage 1 assessment became non-statutory from the 2023/24 academic year and the frameworks carrying that footnote went with it, though the figure has not disappeared from guidance altogether: the DfE's 2023 Reading Framework still attributes it to the Standards and Testing Agency.

That matters for how you use the table above. These are descriptive medians rather than targets. They tell you where children currently are, which is a different question from where they ought to be, and a child sitting on the median is not thereby safe.

Who falls behind

The disadvantage gap is the clearest finding in the report. Children eligible for free school meals read 10 to 15 fewer words per minute throughout primary school, and by the end of Year 6 the median for a child never eligible is 120 words correct per minute against 106 for a child who has been.

Then the finding that should shape how you spend your pupil premium. FFT ran a decomposition analysis on Key Stage 2 reading scores and found that around half of the gap between disadvantaged children and their peers is explained by fluency alone. Half. For children with SEND, where the attainment gap is larger still, fluency explains around two thirds of it.

Gender behaves differently. Girls and boys have almost identical fluency across primary school, yet 78 per cent of girls reached age-related expectations in the 2024 Key Stage 2 reading test against 71 per cent of boys. Whatever drives that gap, it is not fluency.

English as an additional language behaves differently again. EAL children read at broadly the same rate as their peers but score slightly lower on comprehension, which is a useful reminder that words per minute measures how the words come off the page and not whether the child understood them.

The clearest picture of what fluency buys a child comes from grouping Year 6 children into bands. Of those reading below 100 words correct per minute, 42 per cent reached age-related expectations in the Key Stage 2 reading test. Between 100 and 130, that rose to 78 per cent. Above 130, 95 per cent. These are associations rather than proof of cause, but the shape of them is hard to ignore.

What to do with these numbers

Numbers without action are just data.

Screen termly from Year 1. One minute, one unpractised passage, one score, with accuracy and prosody noted separately. It takes two minutes per child and tells you more than most formal assessments. If you have not run a timed reading before, here is a step-by-step guide.

Flag your bottom quartile urgently. At the end of Key Stage 1, below 45 words correct per minute is the line FFT drew. These children are not going to catch up by accident, and the gap compounds every year you leave it.

Track progress, not just attainment. A child reading at 60 who was at 40 last term is moving well. A child reading at 90 who was at 88 has stalled. The trajectory matters as much as the score. At Fetcham, thirty two Year 2 children moved from a mean of 66 to 78 words correct per minute across four months, which told us more about whether the teaching was working than any single termly snapshot did.

Do not use words per minute as a speed target. This is the one that goes wrong most often. If children believe the goal is to read fast, they will trade accuracy and understanding for pace, and you will have made the problem worse while the numbers improve. Frame it as a health check, not a race.

Pair the score with prosody. A child with a strong rate and flat expression needs something different from a child reading slowly but with meaning. The number alone does not tell the whole story.

And teach fluency explicitly. Screening finds the problem. Teaching solves it. Timothy Rasinski's work points consistently in the same direction: children improve when they hear fluent models, practise with familiar, manageable texts and read for a real audience.

Why this matters now

The DfE's Every Child Achieving and Thriving white paper, published in February 2026, announced a statutory reading fluency and comprehension test in Year 8. For the first time, fluency will be formally assessed beyond primary school. Ofsted's 2024 English subject report had already said that too many schools treat phonics as the end of reading instruction rather than the beginning.

For the first time we have English data on where our children actually are. The children in that bottom quartile are already in your building, and most of you could name three of them before you assessed anybody. The numbers only tell you how far behind they have drifted while nobody was counting.

If you want to start screening this term, the simplest approach is a one-minute timed reading with an unpractised passage. Here is how to do it.

---

Correction. An earlier version of this post carried a table of median WCPM scores attributed to FFT Education Datalab. Those figures were not FFT's. They were the Hasbrouck and Tindal (2017) US oral reading fluency norms, mapped grade-for-grade onto UK year groups, which also shifted them a year out of alignment. The bottom-quartile thresholds in the action section came from the same source and were wrong for the same reason. Both have been replaced with FFT's own figures, taken directly from Appendix 2 of the report linked below. Apologies to anyone who used the earlier numbers.

References

FFT Education Datalab (2024). Reading fluency in England: interpreting data from the FFT Reading Assessment Programme. ffteducationdatalab.org.uk

FFT Education Datalab (2024). Measuring reading fluency during primary education. ffteducationdatalab.org.uk

FFT Education Datalab (2024). Low oral reading fluency at the end of Key Stage 1. ffteducationdatalab.org.uk

Standards and Testing Agency. Teacher assessment frameworks at the end of key stage 1. gov.uk

Rasinski, T.V. (2010). The Fluent Reader. Scholastic.

Department for Education (2023). The Reading Framework.

Department for Education (2026). Every Child Achieving and Thriving.