What Is the Best Age to Start 11+ Preparation?

Every page on this question answers with a year group. None of them names a date. For September 2027 entry the registration deadlines in Kent, Buckinghamshire and Sutton all fell inside the summer term of Year 5, months before the exam itself. This page sets the statutory curriculum calendar against the real admissions calendar, and works out what that actually means for when a family should start.

Why is the registration deadline, not the exam, the real gate?

Ask when to start and the internet answers with a year group — Year 4, Year 5, or a vague band of months. Not one of the pages ranking for this question names a single date, and a date is what actually binds a family. In Kent and Buckinghamshire, registration for September 2027 entry closed before the school year containing the exam had even begun. Buckinghamshire's window ran from 1 May to 2 June (3pm) 2026. Kent's opened on Monday 1 June 2026 and closed on Wednesday 1 July 2026. Both of those closing dates sit in the summer term of Year 5. A family that decides in September of Year 6 to start preparing has not started late in those two counties. They have not started at all, because an unregistered child cannot sit the test and there is no route back in.

The exam then follows within days of the new school year. Buckinghamshire ran its Practice Test on 8 September 2026 and the Secondary Transfer Test on 10 September 2026, with results on 9 October 2026. Kent tested on Thursday 10 September 2026 and released results on Thursday 15 October 2026. London runs a little later but the gates are just as hard. Registration for the Sutton Selective Eligibility Test ran from 1 May to 31 July 2026, with an access arrangements deadline of 12 June 2026 and the test itself on 15 September 2026. The Tiffin Girls' School closed registration at 12.00 noon on Tuesday 1 September 2026, the first school day of Year 6, and states plainly that "no late registrations will be accepted". The Henrietta Barnett School's admissions page already reads that applications for the entrance examination in September 2026, for entry in September 2027, "are now closed", adding that it does not accept late applications.

That reordering matters more than any opinion about ideal ages. The question is not really when preparation should start but when the decision must be made, and for Kent and Buckinghamshire families the answer is the summer term of Year 5. Everything else follows from it. If the decision has to be made by early June of Year 5, then the evidence you use to make it — your child's reading, their arithmetic fluency, their school's view — has to be gathered in Year 4 and the spring of Year 5. That is the real reason Year 4 keeps being suggested as a starting point. It is not that a nine-year-old needs to be drilled. It is that the paperwork gate arrives sooner than parents expect. If you are weighing up grammar school entrance preparation for a child now in Year 4, the first job is not buying practice papers. It is finding your authority's registration page and writing two dates in a calendar.

What has your child actually been taught by the time the test arrives?

The national curriculum in England is statutory, and it says in which year each topic must be taught. That makes it the fairest available yardstick for whether a piece of preparation is remedial, on track, or years ahead of a child. Set against it, several of the topics that appear routinely in selective maths papers arrive late. Percentages are a Year 5 requirement: that is the first year the programme of study asks pupils to "recognise the per cent symbol (%) and understand that per cent relates to 'number of parts per 100'". Ratio and proportion is a Year 6 domain, and the Year 5 guidance describes Year 5 multiplication and division work as support "for the introduction of ratio in year 6". Algebra is Year 6 as well, the year in which pupils "should be introduced to the use of symbols and letters to represent variables and unknowns".

Now lay a test date over that. A child sitting the Kent or Buckinghamshire test on 10 September 2026 is a few days into Year 6. Ratio and algebra are Year 6 domains, and the national assessment that examines them comes at the far end of that year: the Key Stage 2 timetable put the tests on 11 to 14 May in 2026, with the maths papers on 13 and 14 May. Mid-May is roughly eight months after a test held in the first fortnight of September. So a selective test in the autumn of Year 6 is not measuring what school has taught. It is measuring the part of Year 6 that has not happened yet.

That cuts both ways, and both directions are worth saying out loud. It is the honest case for structured preparation: no reasonable amount of classroom teaching will have covered ratio or algebra by 10 September of Year 6, so a child who meets those topics for the first time in the exam hall is being tested on unseen material. It is also the honest case against very early drilling. Percentages are a Year 5 topic, so a Year 3 child being drilled on them is working two school years ahead of the statutory programme, usually before the place value and multiplication fluency that make percentages make sense are secure. The Year 4 requirement is the one that actually underpins the maths paper: pupils must "recall multiplication and division facts for multiplication tables up to 12 × 12". Get that solid and most of the arithmetic in a selective paper stops being a bottleneck. Skip it and no amount of ratio practice compensates.

Does the right start date depend on which test you are aiming at?

There is no single right start date, because the tests do not test the same things. Buckinghamshire publishes its weighting: "50% of the standardised score for verbal skills, 25% of the standardised score for mathematical skills, 25% of the standardised score for non-verbal skills". Verbal reasoning and non-verbal reasoning appear nowhere in the national curriculum. Add those two together and 50% plus 25% is 75% of the Buckinghamshire score resting on material no primary school is required to teach. Sutton's Selective Eligibility Test, the shared first-stage test used by a group of selective Sutton schools, goes the other way. It is "made up of two papers; a multiple-choice Maths paper and a multiple-choice English paper", and Sutton Grammar School states that "the tests do not include verbal reasoning or non-verbal reasoning".

The practical consequence is that the same child, with the same ability, has two different correct answers depending on where the family lives. For Buckinghamshire, three quarters of the score sits outside anything the school day will deliver, and unfamiliar question types take time to become familiar. That is a genuine, evidence-based argument for a longer runway. For the Sutton schools, a child who is on track in Key Stage 2 maths and reading widely is already doing most of the work, and what remains is format, timing and multiple-choice technique, which is a much shorter job. Advice that gives one national answer to "when should we start?" is not so much wrong as unanswerable, because it has not asked which test.

Buckinghamshire also makes the target unusually concrete. Rather than a per-school cut-off that moves with the applicant pool, the council states that "children who achieve a score of at least 121 in the test are considered qualified for a Buckinghamshire grammar school", and that score applies across all 13 grammar schools. That is a fixed bar a family can aim at from Year 4 onwards. Kent, by contrast, publishes a threshold that has two parts at once, which we come to below. Before you buy a single practice paper, find out which of these situations you are in. It changes the timetable more than your child's birthday does, and our notes on admissions test preparation set out how the formats differ.

Is there a free readiness check in June of Year 4?

There is a free, statutory, dated readiness check sitting in the middle of Year 4, and almost nobody uses it for this purpose. The multiplication tables check is taken by every eligible Year 4 pupil in England. It is "an onscreen check consisting of 25 times table questions", with three practice questions first and then "6 seconds to answer each question". In 2026 schools had to administer it "between Monday 1 June and Friday 12 June", with a catch-up window from 15 to 19 June 2026. There is no pass mark, which is precisely why it is useful. It is a fluency reading, not a verdict.

What makes it usable is that the national figures are published, so a parent can place their child rather than guess. The average score has risen every year since the check became statutory: 19.8 in 2022, 20.6 in 2024 and 21.0 in 2025. Full marks is the single most common outcome, achieved by 37% of eligible pupils in 2025, up from 34% in 2024.

Read those numbers properly. A national average of 21.0 out of 25 is 84% of the available marks, and the average has climbed 1.2 marks since 2022, because 21.0 minus 19.8 is 1.2. More usefully, 25 out of 25 is the most common single score, so more children get everything right than get any other result. For a child aiming at a selective test two years later, full marks in June of Year 4 is therefore not an exceptional outcome to celebrate. It is the ordinary one, and it is the sensible internal target. A child scoring in the high teens is not failing, but they are carrying an arithmetic tax into every timed maths paper they will ever sit, and that is the specific, fixable thing to work on in Year 5. Not percentages. Not algebra. Tables.

Does a summer birthday change when you should start?

Summer birthdays generate more anxiety on parenting forums than any other single factor here, and the data lets us answer in two halves rather than reassure vaguely. The classroom gap is real and still measurable at the end of Year 4. In the 2025 multiplication tables check, August-born pupils averaged 20.4 and September-born pupils averaged 21.6. That is a gap of 1.2 marks out of 25, since 21.6 minus 20.4 is 1.2, between children in the same year group, at exactly the age when preparation is usually being considered. Both groups improved on the year before — August-born from 20.0 to 20.4 and September-born from 21.3 to 21.6 — but the gap did not close.

The second half is the correction. Kent states that in its test "a slight adjustment is made to take account of each child's age so that the youngest are not at a disadvantage". Buckinghamshire standardises using "each child's age in years and months on the date they take the test", and says the point is to make sure "that no child is unfairly advantaged or disadvantaged because of their age". Not by term. By month.

Put the two halves together and you get an answer that neither reassures nor alarms. Nothing in that data says a summer-born child is scored more harshly, because the scoring system already compensates, to the month. What it says is that at the point the decision is being made, a summer-born child is on average slightly further back along a curriculum they have had fewer months to absorb. So an earlier start for a summer-born child is justified, but for the right reason: closing a real curriculum and confidence gap, not out-running a scoring system that has already accounted for it. If anyone sells you extra tuition on the basis that summer-born children are penalised in the test itself, that premise is wrong in both counties.

How long should 11+ preparation actually last?

The strongest claim on most pages about this question is that starting too early causes burnout. It is usually asserted and never sourced. The Education Endowment Foundation's Teaching and Learning Toolkit lets us say something sharper and evidenced. On one to one tuition it reports an average of five months of additional progress, and it is specific about the shape: "Short, regular sessions (about 30 minutes, three to five times a week) over a set period of time (up to ten weeks) appear to result in optimum impact." Up to ten weeks. Not up to three years.

The second finding should change how families spend the time they do commit. The toolkit reports that "effects in mathematics appear to be substantially lower (+2 months) than in literacy (+6 months)". Six divided by two is three, so the measured effect of tutoring in literacy is three times its measured effect in maths. Small group tuition shows four months on average, with a clear size cliff, because "once group size increases above six or seven there is a noticeable reduction in effectiveness", and it shows the same roughly ten-week optimum block.

Two independent things now point the same way. The research says language and reading are where tutoring moves the needle furthest, and Buckinghamshire says verbal skills are half the score. A family with limited time and a child who reads reluctantly should be spending that time on reading, vocabulary and comprehension, not on a fourth year of arithmetic worksheets. One honest caveat belongs here: the toolkit's figures are averages from school-based tutoring aimed at attainment generally, not from selective-entry coaching, and nobody has published a controlled trial of grammar school entrance preparation. Treat the ten-week block as a shape rather than a prescription. It is still a far better guide than an unsourced assertion, and it argues squarely against multi-year drilling. Intensity late beats duration early.

Should you be preparing at all?

Kent County Council is unusually blunt about who its grammar schools are for. "Many children will find the test difficult," it says, and its grammar schools "provide for children in about the top 25% of the ability range". That is not discouragement, it is arithmetic. Most children who sit the test will not be assessed as suitable, and a family deciding in Year 4 whether to commit two years to this deserves to know it before they start rather than after.

The published threshold shows how the bar is actually built. For the 2025 Kent Test, "children needed a total score of 332 or more, with no single score lower than 108", where "test scores range from 69 to 141" and "the highest possible total score is 423". Two things fall out of those figures that are worth working through. First, 423 divided by 141 is 3, so the total is three standardised scores added together. Second, and more usefully, clearing the individual floor on all three papers gives 108 times 3, which is 324, and that is eight marks short of the 332 total. A child cannot qualify by scraping the minimum everywhere. And 332 out of 423 is 78.5% of the maximum, spread across every paper rather than concentrated in one.

That is the practical reason a lopsided preparation plan fails. A child who is very strong in maths and weak in English cannot buy their way past a per-paper floor with a maths score, because the floor is checked separately on each paper. Buckinghamshire's single qualifying score works differently but rewards the same balance, because half of it is verbal. So the honest test of whether to prepare at all is not whether a child is clever. It is whether they are within reach across every paper, including the ones they like least, and whether the family is willing to spend the summer term of Year 5 on it. If the answer to either part is no, the kinder decision is to aim elsewhere, and a strong comprehensive with good subject tutoring alongside it is not a consolation prize. Some families also find that senior school pre-tests suit their child's timing better.

Year by year: what to do, and when?

Here is the same advice with dates attached instead of adjectives. In Year 4, the fixed event is the multiplication tables check in the first fortnight of June. Use the result. It is free, it is national, and it tells you whether arithmetic fluency is a problem before anything else is. Nothing else in Year 4 needs to happen. Reading widely, talking about words, and being fluent in tables is the whole job at that age, and it is the job the curriculum has already set.

Year 5 is where the calendar takes over. Registration opened on 1 May 2026 in both Buckinghamshire and Sutton and closed between 2 June and 31 July 2026 depending on the authority, while Kent's ran from 1 June to 1 July 2026. Put those in a calendar in the autumn of Year 5, not the summer. This is also the term structured preparation belongs in, and Kent tells you what shape it should take: each section of its test is "a 5 minute practice exercise followed by a 25 minute test", the English and maths paper "will take 1 hour", and the writing exercise runs 40 minutes including 10 minutes of planning. That is the real endurance requirement. Working in 25 minute focused blocks matches the test and matches the toolkit's session shape at the same time. Weekend mock marathons match neither.

Year 6 barely exists as a preparation window in the selective counties, which is the whole point of this page. Buckinghamshire states that "most children will take the Secondary Transfer Test during the autumn term of Year 6", and it gives every eligible child a familiarisation booklet and a practice test. Use those first, because they are official, free and format-accurate, and they come before any decision about paid help. For families whose children are in Year 4 right now, Buckinghamshire has already published the next cycle: registration opens early May 2027 and closes early June 2027, with tests in early September 2027, for "children born between 1 September 2016 and 31 August 2017". If that is your child's birth window, you now know your own deadline, and you can plan backwards from it instead of forwards from a guess.

Frequently Asked Questions