What last summer's exam results actually tell us — and what they mean for the next few weeks

Exam season is approaching again, which means that within a few months, a fresh set of headline grade percentages will land, get repeated for about thirty-six hours, and then vanish from the conversation until the same week next year. Before that cycle starts again, it's worth pausing on what last summer's results actually showed — because the numbers underneath the headline are more useful, and more stable, than the yearly "up or down" framing usually suggests.

The headline numbers, briefly

Last August's results showed a little under seven in ten GCSE entries achieving a grade 4 or above — the standard often described informally as a "pass" — and a little over one in five reaching a grade 7 or above, the rough equivalent of the old A grade. At A-level, around one in eight entries reached the very top grade. None of those figures moved dramatically from the year before. They sat close to where recent years have settled, after the larger swings that followed the pandemic-era assessment changes.

That's it. That's the headline. It's accurate, and it's almost entirely useless on its own — because a single national percentage tells you almost nothing about the experience of any individual student, class, or school.

Why the national number isn't the useful number

Here's the thing about a national grade distribution: it's an average of an enormous number of very different stories, compressed into a handful of digits. A national figure moving by half a percentage point can mean almost anything at the level of an actual classroom — it can hide a cohort that did notably better, another that struggled with a specific paper, a school that made real strides with a group it had been worried about, another that's still working out why a topic didn't land the way it was taught. All of that detail gets averaged away before it ever reaches a headline.

Which means the genuinely useful exercise — the one worth doing in the weeks after results land, and arguably worth starting before they do — isn't reading the national percentage. It's looking at your own data, at the level where it actually means something: which topics, which question types, which groups of students moved the way you expected, and which didn't.

Turning results into something you can act on

The real value of exam data isn't in knowing where the country landed. It's in knowing where your students landed relative to where you expected them to be — and specifically, in catching the gap between "taught" and "understood" before it hardens into a result. That's most useful when it happens continuously through the year rather than once in August, which is exactly the gap that something like exit tickets is built to close: a fast, low-friction read on the whole room, often enough that a misconception surfaces while it's still small, rather than showing up — fully formed — in a summer grade distribution that's already too late to do anything about.

The national headline will arrive again in a few months, say roughly the same thing it said last year, and move on within a day or two. The detail worth building a year's teaching around was never really in that number. It was always in the smaller, more specific picture — the one that's actually visible to you, considerably earlier, if you're set up to look at it regularly rather than waiting for August to tell you what already happened.