Everything you need to know, in one place, with the sources shown.

What a grammar school is, in law

State-funded, locally maintained, not locally controlled, and free to refuse you.

In England

163

In Wales

none

In Northern Ireland

65 or 69

A grammar school is a state-funded school that selects all of its pupils by academic ability. There are 163 in England, listed by statute under the Education (Grammar School Ballots) Regulations 1998. There are none in Wales, which closed or converted them in the 1970s, and none in Scotland in this sense, though a few historic schools keep the name. Northern Ireland has a substantial separate system, and its grammar schools work differently: since the 1989 open enrolment reform they must take pupils up to capacity rather than selecting to a fixed number. The number of Northern Irish grammar schools is genuinely disputed between sources, which say 65, 67 or 69.

The legal position confuses parents, so it is worth stating plainly. A maintained grammar school is funded and infrastructure-supported by the local authority but not controlled by it. The authority has a duty to fund the school; the school's own governors decide who gets a place. The admissions rules are set by the governing body, which is what the law calls the admission authority. The statutory basis is the School Standards and Framework Act 1998.

Two consequences follow that most families do not expect. First, selection is protected: only designated grammar schools may select their entire intake on academic ability. Second, and more surprisingly, they do not have to fill their places. Under the School Admissions Code, a grammar school that offers nobody a place it should have is acting lawfully. There is no obligation to offer a place to a child who meets the standard if the school has decided it wants fewer pupils.

One historical detail that explains the geography. Almost all the English comprehensives of the 1960s and 70s came out of grammar-school conversions, leaving a small residue where ending selection was tried. Only one ballot to abolish selection has ever been held, at Ripon Grammar School in 2000, and parents rejected the change by two to one. That is why the 163 schools are clustered in particular counties and local authorities rather than spread evenly: only 35 local authorities have any grammar school at all, and around 60% of them sit in just eleven.

A source of confusion worth flagging: many English grammar schools have since converted to academies. An academy grammar school is no longer council-maintained, but it still selects. So the phrase local authority grammar school is not quite the same set as the 163.

• The campaigning body is the National Grammar Schools Association. There is no British Grammar School Association, and the Grammar Schools Association does not exist under that name either.
• The NGSA's own website has not been updated since 2017 and still claims 164 schools, which is why the statutory figure of 163 is the one used here.

How your child actually gets a place

One hard national date, one ranked application, and a set of rules schools must follow when places run out.

Everything is governed by the School Admissions Code 2021, statutory guidance in force since September 2021. There is one hard national date in the whole process: 31 October, the national closing date for secondary school applications. For a September 2026 start that meant applying by 31 October 2025. Your local authority runs a single co-ordinated application in which you rank your preferences, usually three, though London authorities generally allow up to six.

How the place is then decided varies, and the variation matters. Where a grammar system survives across a district, the test is used to identify a subset of children and the remaining criteria allocate places among them if too many qualify. In an isolated grammar school, places go in rank order of test performance alone. The Code is deliberately not prescriptive and offers no definitive list of acceptable oversubscription criteria, but the common ones are a sibling already at the school, distance from the school or from a named measurement point, catchment areas, feeder schools, and social or medical need. A school must state precisely how it defines home and from which point it measures distance.

One rule catches parents out: a school may not charge for the process. Not for tests, not for contributions, and not even for a refundable deposit. A school asking you to pay to sit its entrance test is acting outside the Code.

If you are refused, appeals are governed separately, by the School Admission Appeals Code 2022. You must be given at least 20 school days from notification of refusal to lodge an appeal, at least 10 school days' notice of the hearing, and a normal-round appeal must be heard within 40 school days of the deadline. Every authority must publish its appeals timetable by 28 February each year. If a panel upholds your appeal the school must admit your child. You can also ask for an interim decision, an early place while the appeal runs, if you have applied elsewhere and your child would otherwise be without a school.

How competitive a place really is

The national offer rate is reassuring in a way that is misleading, and no waiting-list figures are published.

First choice, all secondary

83.6%

One of their choices, all secondary

96.4%

Local authorities with any grammar school

35

For September 2026 entry, 83.6% of secondary applicants got their first choice and 96.4% got one of their ranked preferences. Those are the official figures, and they are true. They are also close to useless for a family applying to a grammar school, because they are dominated by the non-selective schools where the great majority of applicants go.

The DfE does publish a breakdown by whether an authority has selective schools, but we could not retrieve it, so the honest answer is that we cannot give you a grammar-school-specific offer rate. The closest thing to a comparable figure came from Schools Week in September 2026, reporting that free school meal rates at the average grammar were 9.2 percentage points lower than elsewhere in their own catchment areas in 2024. A catchment where the grammar draws noticeably fewer disadvantaged pupils is not a catchment where the test is doing much work.

On waiting lists, we have to be blunt: we could not find a single authoritative, current, published count of grammar-school waiting-list lengths anywhere. You will see numbers quoted. Treat them as unverified. What can be said is structural: only 35 local authorities have any grammar school at all, and around 60% of them are in eleven authorities, so in most of the country a place at the nearest grammar is not a competitive exercise at all, while in a handful it is the most competitive thing a family will ever do.

• The 83.6% and 96.4% figures cover all secondary applicants. Do not read them as grammar school statistics.
• No authoritative current published count of grammar-school waiting lists could be found.
• In Gloucestershire about 2,500 children a year sit for seven grammar schools, which is a reasonable illustration of the ratio in a genuinely competitive area.

The timeline, with dates

The part you can plan around. Only some of these dates are national; the test date is set locally.

The 28 February date is the one to put in a diary now, even if you are years away. Every authority has to publish its appeals timetable by then, and it is the document that tells you what to do if something goes wrong. Reading it in advance rather than in a panic afterwards is the whole point of it being published months ahead.

One trend worth noticing: test dates are moving earlier, out of the summer holidays and into the summer term of Year 5. Gloucestershire has committed to this, and reporting in 2026 suggested other areas were considering it too, the stated reason being that children lose focus over the summer. If your child is three or more years out, expect the date you are planning around to shift.

• Entry later than 11 does happen, but through late transfer rather than the normal round. Cambridge sells a late-transfer version of its assessment designed for Years 8, 9 and 10, roughly ages 12 to 14, and admissions authorities run their own tests for mid-year entry. It is per-school and competitive on its own terms.
• The overwhelming norm is entry at 11, into Year 7.

What changed in 2026, and what did not

Free school meals widened substantially. The admissions cycle did not move at all.

From the start of the 2026 to 2027 academic year, all children in households receiving Universal Credit are entitled to a free school meal in maintained schools, academies and free schools, which includes maintained grammar schools. This was made a legal requirement by the Children's Wellbeing and Schools Act 2026. There are now two categories. Targeted free school meals cover Universal Credit households with annual earnings of 7,400 pounds or less and keep the associated pupil premium and disadvantage funding. Expanded free school meals cover Universal Credit households above that figure and are meals only. Transitional protections ended at the end of 2025 to 2026, so eligibility is now reassessed annually, and you still have to apply.

This matters for admissions rather than for lunch. Pupil premium and free school meal status feature in the oversubscription criteria of some authorities, and Buckinghamshire goes further, recording that some of its grammar schools have a limited number of places for children in receipt of pupil premium who achieve a certain score, at a slightly lower threshold than the usual qualification score. That is a genuine, checkable adjustment, and it is not available everywhere.

Gloucestershire has made the most significant change of any single authority. Its seven grammar schools appointed the not-for-profit Future Stories Community Enterprise to run their Year 7 entrance test for entry from 2028, first sitting in summer 2027. The test moves from September of Year 6 to the summer term of Year 5, is based on the national curriculum, and removes non-verbal reasoning, on the stated grounds that it is the component that typically features in private tutoring.

Expansion is happening school by school rather than through any national programme. We looked for a Grammar School Expansion Programme on GOV.UK and could not find one under that name, so we are not going to describe one. What is verifiable is individual expansions: Oakwood Park Grammar School in Maidstone had a 7.3 million pound expansion approved, taking Year 7 places from 160 to 190 from September 2027,. In Maidstone, about 55% of Year 7 grammar places are for girls and 45% for boys.

And the thing that has not changed is the thing most families plan around: the School Admissions Code 2021 remains in force, 31 October is still the closing date, and National Offer Day is still 1 March or the next working day. The 2026 cycle ran to those dates.

A plain-English glossary

The words that turn up in admissions paperwork and prospectuses, defined.

Most of the confusion around the 11+ is vocabulary rather than substance. These are the terms that turn up in prospectuses, admissions emails and conversations with other parents, defined as plainly as the sources allow.

This is the one place on the page where we would rather be short than thorough. A term can mean different things to different providers, and where that is true we say so rather than picking one meaning and hoping.

Registering and applying are different jobs

A genuinely common mistake, and the one deadline families miss.

This trips up more families than anything else on the page, so it is worth saying on its own. Registering for a test and applying for a school place are two separate processes, usually with two separate deadlines, and doing the first does not do the second.

Registering is what you do with the exam provider or the admissions authority to get a test. Applying for a place is what you do through your local authority's single co-ordinated form, where you rank your preferences, by 31 October. Miss the registration deadline and there may be no test to sit. Miss the application deadline and the test result has nowhere to go. The second is national and fixed. The first is local and often much earlier.

The order is usually registration first, then the test, then results, then applying. In the 2027 cycle, Warwickshire opened registration for the test on 8 May 2026 and Lancaster Girls' Grammar School closed registration on 9 September 2026, while the national application deadline was 31 October 2026. That gap is about seven weeks, and it is not a gap any of those schools will chase you about.

Two more mechanical points. A school may not charge you for the process, including for tests and even for a refundable deposit, so a school asking for payment to sit its entrance test is acting outside the Code. And if you are refused, you have at least 20 school days from the notification to lodge an appeal, with the authority's timetable published by 28 February.

• Registering for a test is not applying for a place. They have different deadlines and one does not do the other.
• The registration deadline is local and often much earlier than 31 October.
• The application deadline is national: 31 October, every year, no exceptions.
• No school may charge for tests, contributions or even refundable deposits.

The four components, and what each one is

What a child is actually being asked to do in each of the four papers.

Not every test has all four, and this is the first thing to get straight: which components your child sits, and in what balance, is decided by the school or the local authority, not by you. GL covers five areas across ages 7 to 14: verbal reasoning, non-verbal reasoning, spatial reasoning, English and maths. Cambridge Select Insight covers three: numerical, verbal and non-verbal ability. Ask your school which of its own components apply before you buy anything.

Verbal reasoning is the one most people mean when they say reasoning, and it is not reading comprehension. It is reasoning with words: following an argument, spotting which of two statements follows from a set of rules, working out what a word means from how it is used. GL's description is that these tests identify pupils whose strengths lie in English, history, languages and the arts, and that they are less reliant on curriculum exposure or rehearsed content. That last clause is the important one: it is designed so that having done the syllabus does not hand you the paper.

Non-verbal reasoning is reasoning with shapes and patterns rather than words, and GL describes it as less reliant on language, curriculum exposure or rehearsed content. It is the component Gloucestershire is removing from 2028, on the stated grounds that it is the one that typically features in private tutoring. So it is the component most exposed to commercial preparation, and the one the profession is moving away from.

English is a curriculum paper, and the content ceiling matters. GL says its English and maths papers are based mainly on curriculum content reached during the early part of Year 6, and elsewhere says up to the end of Year 5. This is the component where a child's school work is most directly the preparation, which is why a family whose child reads widely has already done a good deal of it without knowing.

Maths is the other curriculum paper, and it is the one the evidence says tutoring helps least with. The EEF finds the effect of one-to-one tuition in maths is about two months against about six in literacy. It is also the component with the least room for reading, so a family with a strong reader and a weaker maths child should be clear that the two problems need different answers.

Spatial reasoning appears in GL's list of areas but is not one of the four standard components, and is worth being careful about. If a school mentions it, it is a bespoke paper, and the specification will be the school's own.

What is actually sat

There is no single 11+ in England. Not one paper, not one provider, and not regulated by any national body.

The single most useful thing to understand about the English 11+ is that it does not exist as a national exam. It is not regulated by Ofqual or any national exams body, there is no common format, and no two authorities sit the same paper. What exists is a market of providers, and a patchwork of local decisions about which of their papers to use.

GL Assessment describes itself as the largest provider of 11+ tests to schools in England, delivering over 300,000 papers a year, covering ages 7 to 14 across verbal reasoning, non-verbal reasoning, spatial reasoning, English and maths, on paper, multiple choice, thirty minutes upwards. Cambridge, trading as Cambridge Select Insight, sells a digital assessment averaging sixty minutes with three sections: numerical, verbal and non-verbal ability, for entry into Years 7 to 10. Schools' own and not-for-profit providers are now a real part of the market: the not-for-profit Future Stories Community Enterprise, founded by Reading School, provides tests for 12 grammar schools with more than 8,000 pupils sitting its papers each year, and the not-for-profit Consortium for Selective Schools in Essex provides the same test to 10 member schools.

GL's own explanation of why the papers differ is the important one, and it is worth reading in full: tests vary because local authorities or individual schools decide which subjects to test, as well as the balance of question types, and an authority or school may add its own questions to those from a provider. In other words the test is designed locally, by the school your child is applying to.

A common parental belief is that most papers contain a verbal reasoning test because it is the best predictor of future potential. That claim comes from GL, and GL is the commercial supplier of verbal reasoning, so it is interested testimony. The underlying research is real; the emphasis belongs to the seller.

• GL: the largest 11+ provider in England, 300,000+ papers a year, ages 7 to 14, paper and multiple choice, 30 minutes upwards.
• Cambridge Select Insight: digital, about 60 minutes, numerical, verbal and non-verbal sections, for Years 7 to 10.
• FSCE: a not-for-profit founded by Reading School, testing for 12 grammar schools and 8,000+ pupils a year.
• CSSE: a not-for-profit running since 1993, serving 10 Essex member schools.
• Cambridge states it does not sell practice materials and does not endorse commercial resources, on the grounds that this makes the test harder to tutor to.

What a score actually means

A standardised score says how your child did against others of the same age. It is not a measure of ability, and it does not travel between schools.

A raw mark, meaning the number of questions right, is close to meaningless on its own. A paper sat by Year 5 children in June is a different paper-sitting from Year 6 children in September, so the providers convert raw marks into a standardised score adjusted to what an average child of the same age and month would have achieved. The scale is centred so that 100 is average, and 115 is one standard deviation above average.

The consequence parents most often miss is this: a standardised score is a statement about your child relative to other children of the same age. It is not an absolute measure of ability, and a score from one school cannot be compared with a score from another. Two children with the same standardised score at two different schools have not necessarily done equally well, because the two papers are not the same paper.

You will usually not be given raw marks. Reports indicate that raw marks are not disclosed by all schools, and parents are given Standard Age Scores instead. There is a checkable oddity here, and it is worth knowing: asked under the Freedom of Information Act for its 11+ data, Altrincham Grammar School for Boys replied that its exam provider is not subject to the Act, as it is not a public body. So some of the detail families would find useful is not obtainable from either the school or the provider.

GL publishes no conversion table and no worked example for its 11+ papers. It describes the standardisation process, and it points parents asking about format to their admissions department. If you need the specifics, your school is the only route to them.

• 100 is average. 115 is one standard deviation above.
• GL's public guide explains standardisation for its school tests such as CAT4, not for the 11+ papers, and gives no 11+ conversion table.
• GL declines to state the format of the 11+ at all: contact your local grammar school or local authority.

What is genuinely free, and what is advertising

An audit, because this is where parents are most often misled.

There is a widely repeated claim that there are a hundred free 11+ practice papers available. We could not find a primary source for it. GL never makes that claim; GL says it offers ten hours of free familiarisation materials. Treat the hundred-paper figure as unverified.

What is real falls into three categories, and telling them apart matters more than any single resource.

Genuinely free, with no strings: GL publishes six ZIP downloads on one page with no email capture, covering verbal reasoning, non-verbal reasoning, English, maths, non-verbal reasoning with maths, and verbal skills. Between them they contain twelve test booklets, answer sheets and a parent's guide for each. ISEB publishes a free walkthrough and several free PDFs including a test framework.

Free but gated: Bond, published by Oxford University Press, offers hundreds of free practice questions, but you have to log in to a Bond account to reach them. Free in the way that a supermarket sample is free.

Advertising for something else: this is the category to be most careful with. ISEB's own guidance on third-party preparation platforms is blunt: in all but one case ISEB has not worked with these companies and has not supplied them any questions, and ISEB cannot guarantee the accuracy of any other third-party platform. Exactly one platform is ISEB-endorsed.

Two caveats on GL's genuinely free material, both read from the documents themselves. The print dates run from June 2017 to January 2024, so the newest free pack is over two years old. And GL restricts redistribution: private tutors and other organisations may not feature the page or copy, reproduce or distribute the materials. That restriction is why you will find a lot of the free papers circulating as downloads elsewhere rather than as links to GL.

What the paid material costs

Real prices, and what the largest single spend reported in the press was.

A single GL practice pack

£9.99

All 12 GL packs

£82.50

ISEB Pre-Tests prep pack

£149

Reported top-end spend

about £600 a year

Those are the prices, and they are not extreme. The figure that should make a parent pause is the one from the press: an investigation in September 2026 found the most expensive preparation costing about 600 pounds a year, with GL bundles at 99 pounds, one commercial package at around 670 pounds a year, and the Essex consortium's packs at 19 pounds, noting that the consortium's 2024 pack was free for children on free school meals but families must pay for more recent years.

The head of Reading School put the reason plainly. Selling practice papers to families creates a tutoring arms race which is unfair upon the most disadvantaged, and the 11+ must not be a high-stakes assessment which needs years of tutoring and hours of practice papers. He is the person whose own school sells papers through the not-for-profit it founded, so it is worth reading his position as a caveat rather than a verdict.

Families interviewed by the BBC in Gloucestershire, after that county changed its test, said the things parents actually feel. One: I feel guilty that we are giving my daughter an advantage over other people because we're paying for her to be tutored. Another: I think we probably will still tutor because everyone else is. And in counterpoint: it's really expensive and we know parents who paid for tutoring for years and then their children didn't get in.

What the evidence actually says about practice

Be clear about the quality of this evidence, because it is weaker than anyone selling you a course will tell you.

We found no controlled research quantifying how much practice raises an 11+ score, and no authoritative guidance on the best age to start. What follows is provider design claims, practitioner argument and reported family behaviour. If you are about to spend money on the strength of a number, this is the number you should look at, and the honest answer is that it is not the one the courses advertise.

On tutoring generally, the Education Endowment Foundation reports a headline effect of about five months of additional progress for one-to-one tuition, described as a moderate impact for moderate cost based on moderate evidence, from 123 studies, last reviewed in July 2021. In primary schools the effect is larger than in secondary, and in maths it is much lower than in literacy. The recommended shape is specific: about thirty minutes, three to five times a week, over a set period of up to ten weeks.

The DfE's independent evaluation of the National Tutoring Programme, Year 3, is a much more sobering document, and it is about the real world rather than a trial. It found that although many results reached statistical significance, the effect sizes were small and equated to one month's additional progress or less. Two years on, at Key Stage 4, effects equated to less than one month's additional progress, and it detected no other longer-term benefits.

The EEF attaches a caveat that matters here more than most of its caveats: evaluations conducted by organisations connected with the approach, for example commercial providers, typically have larger effects, which may influence the overall impact. In other words, if a tutoring company published the research, expect a bigger number.

The exam boards are consistent about what they built their tests to resist. GL says its tests are designed to be less influenced by intensive or repeated preparation, and that the reasoning components are less reliant on curriculum exposure or rehearsed content. Cambridge says it introduces new assessment paradigms to reduce the predictability of assessment content. ISEB says adaptive testing means children do not spend time on questions that are too hard or too easy, and states flatly that the Pre-Tests are based on the national curriculum so additional preparation is not a requirement.

• EEF, one-to-one tuition: about five months' additional progress, moderate evidence, 123 studies.
• EEF, individualised instruction: about four months, based on limited evidence, 198 studies.
• DfE, National Tutoring Programme Year 3: one month's progress or less at Key Stage 2, and less than one month two years later.
• The EEF's caveat: commercial-provider evaluations report larger effects and may skew the overall figure.
• No controlled study of how much 11+ practice helps was found, by us or, as far as we can tell, by anyone else.

State or independent: decide this first

Both take boys, both are called grammar schools, and they run different exams.

This is the single most common source of error in the whole subject, so it comes before anything else. A state grammar school is a selective school maintained or academy-converted by the state, and it is free. A selective independent school is fee-paying. Both take boys, at 11+ or at 13+, both get called grammar schools by parents, and they run entirely different exams with different deadlines and different fee structures.

Confusing the two wastes months. A family preparing with 11+ practice papers for a 13+ independent school is preparing for the wrong exam, and a family who thinks a super selective school is subject to the national admissions code is waiting for an offer that never comes that way.

The two real campaigning and representative bodies are the National Grammar Schools Association, which is parent- and governor-backed, and the Grammar School Heads Association, which represents heads. There is no British Grammar School Association. We checked, and so did two of the people researching this page, independently.

On the size of the boys' sector, the honest position is that the published figures do not reconcile. The House of Commons Library, drawing on DfE pupil statistics for January 2022, gives 163 grammar schools attended by around 188,000 pupils, which is 5.3% of state-funded secondary pupils, concentrated in 35 local authorities with about 60% in just 11. The National Grammar Schools Association says England has 164, on a page whose most recent content dates from 2017. Counting the statutory list by hand gives 56 boys-only, 60 girls-only and 46 mixed, which does not add up to either published total. Treat the proportion as fair and the absolute number as uncertain: boys' schools are roughly a third of the sector, girls' a little more, and mixed around a quarter.

One point that gets assumed and is not supported. Many people assume boys' grammar schools mostly select at 13+, because that is true of the independent tier and is why ISEB runs a full Common Entrance system. It is not true of the state sector. A large group of state boys' grammar schools still select at 11 through the local test, Tunbridge Wells Grammar School for Boys among them. We could find no reliable published statistic for the split, so the page does not give you one.

• State grammar school: free, maintained or academy-converted, selects at 11, run under the national admissions code.
• Selective independent school: fee-paying, usually selects at 13, its own admissions process and deadlines.
• The bodies that exist: the National Grammar Schools Association, and the Grammar School Heads Association. There is no British Grammar School Association.

Boys' schools do not all test the same thing

Two named schools, two completely different papers, and the pattern repeats across the sector.

This is the point that catches families who have researched one boy in one year and assumed it generalises. Sevenoaks selects at 13+ by ISEB pre-test, a reference from the current school, a competitive examination in English and maths, and a group discussion activity. No science, no humanities, and a discussion rather than a one-to-one interview. Royal Grammar School, Guildford has all boys sit a single day of English at one hour fifteen, mathematics at one hour, and verbal reasoning at fifty minutes. Those are two of the best-known boys' schools in the country, and a child could prepare for one and be entirely unprepared for the other.

In the state sector the variety comes from consortia rather than individual schools. The Kent Test covers Kent and Medway, Buckinghamshire runs its own Secondary Transfer Test, the Consortium for Selective Schools in Essex serves ten member schools, and Northern Ireland uses SEAG. A family in one of those areas should find out which consortium serves their target school before buying a thing.

GL, which describes itself as the largest provider of admissions tests to both independent and grammar schools in the UK, covers five areas: verbal reasoning, non-verbal and spatial reasoning, English and maths, on paper, multiple choice, thirty minutes upwards. It claims a bank of 20,000 trialled questions. ISEB runs two separate systems that are easy to confuse: the four adaptive online Pre-Tests, taken once and shared across every school that accepts them, and Common Entrance at 13+, which many independent and some grammar schools use to confirm a deferred place. Confirmed Common Entrance 13+ dates for the current cycle are 2 to 6 November 2026, 25 to 29 January 2027, and listening and speaking from 17 to 22 May 2027 with written papers 7 to 11 June 2027.

What a provider will tell you that is more useful than a weighting is what their test is not measuring. GL says its reasoning components are less reliant on curriculum exposure or rehearsed content, and that non-verbal reasoning helps identify pupils whose future potential is not always reflected in current attainment. ISEB's Pre-Tests are adaptive, meaning the algorithm reassesses after each response and picks the most suitable question, so a child is not spending time on questions far too easy or too hard. The design intent is consistent: the reasoning papers measure how a child thinks, the English and maths papers measure what they have been taught.

Boys' timelines, and what a threshold actually means

Two routes at 13+, one at 11, and a number that is routinely mistaken for a pass mark.

A 13+ place usually has two routes, and they are not interchangeable. Early Decision means registering three years ahead, in Year 6, sitting the ISEB pre-test in the autumn of Year 6, then an assessment day in early January. Later Decision means registering by 1 November of Year 7, sitting the pre-test that autumn, and second-stage assessments in March or April of Year 7. Sevenoaks is currently taking applications for September 2028 and 2029, and has closed for 2026 and 2027. Note that its own website gives two different registration deadlines for the same route, 15 September on one page and 21 September on another, both currently live. That is worth knowing before you rely on a date.

For an 11+ place at a state boys' grammar school the local authority runs the test. At Tunbridge Wells Grammar School for Boys the supplementary information form for pupil premium-eligible candidates is due on 31 October, offers arrive on National Offer Day in early March, and for September 2026 entry the Kent Test threshold was 332 overall with no individual paper below 108.

That number needs a warning, and the school itself supplies it: it is not a criterion for a place at that school. A threshold across a whole testing area is not the same thing as the score at which a particular school would make you an offer, because where you live and which paper you sat determine which places you are competing for. Reading a threshold as a target is one of the ways families end up spending money on the wrong thing.

On when to start, the honest answer separates two things that are usually run together. Registration is earlier than most parents expect, because the 13+ Early Decision route wants you three years ahead. Academic drilling is explicitly optional per ISEB and explicitly something GL designs against. Register early. Prepare calmly.

• 13+ Early Decision: register about three years ahead, in Year 6, pre-test that autumn, assessment day in early January.
• 13+ Later Decision: register by 1 November of Year 7, pre-test that autumn, second stage in March or April of Year 7.
• 11+ state route: the local consortium test, supplementary forms due 31 October, offers on National Offer Day in early March.
• A published threshold across a testing area is not a pass mark for a particular school.

What the exam board tells parents to do

In the words of the person who runs the qualifications, which is the most direct advice available.

The most direct statement of official advice we found comes from ISEB's Head of Qualifications, Hayden Wood, writing in Education Choices Magazine in autumn 2025 and republished by ISEB that October. Two of his points are worth quoting rather than paraphrasing. The second is: trust the school's preparation, extra practice is optional. The seventh is: focus your preparation on building confidence, not pressure. He also recommends applying early for any special needs adjustments, using only ISEB-endorsed resources, taking the free platform familiarisation, and prioritising sleep, exercise and a balanced routine.

That is advice from the body that sets the exam, published under its own name, and it is worth weighing against anything a course sells you. It is also consistent with what the test designers say. GL says its tests are designed to support the requirements of the Equality Act and to be less influenced by intensive or repeated preparation.

The documented errors that follow from this are the predictable ones: mistaking the 11+ route for the 13+ route or the reverse, registering late, preparing with the wrong provider's papers, and converting preparation into pressure. We could not find credible independent research quantifying these failure modes, and the pages that rank for 11+ preparation mistakes are almost entirely tutoring companies selling courses, which we have not treated as evidence.

One thing we would add, which the exam board does not say because it is not their job. The single best preparation available to a child is a parent who knows what the process actually is. The families who understand that a threshold is not a pass mark, that registration and application are different jobs, and that the same test is sold under two names for two different ages, do better in the room than the families who have bought more practice material. That is not a finding, it is an observation, and we would rather label it as one.

• ISEB's Head of Qualifications: trust the school's preparation, extra practice is optional.
• ISEB's Head of Qualifications: focus your preparation on building confidence, not pressure.
• ISEB also advises applying early for adjustments, using only endorsed resources, and prioritising sleep and exercise.

What policy is actually doing

Including one checkable fact about the white paper that surprised us.

Free school meals widened in September 2026. From that date all households receiving Universal Credit qualify, including those above the previous 7,400 pound income cap, with savings of up to 495 pounds a year per child. Over half a million additional pupils became eligible. The detail families miss is that eligibility is granted automatically but families still have to apply, and that free school meal status keys off disadvantaged-priority criteria and pupil premium supplementary forms at some grammar schools, so it is an admissions fact as much as a lunch one.

The February 2026 schools white paper, Every child achieving and thriving, was first published and laid in Parliament on 23 February 2026 and updated on 27 April. Its full text contains the words selective and grammar zero times. We checked, because it is the obvious thing to assume had changed. It does propose consulting on School Admissions Code changes, including tightening the rules that ensure banding produces representative intakes and improving the information available for in-year admissions. Both of those would matter to a family applying for a grammar school place, and neither is settled.

The expansion route is still expansion rather than new schools. The general prohibition on selection remains; grammar schools with arrangements dating from 1997 and 1998 are the historical exception, and an existing grammar may expand only as a change to an existing school. The Selective Schools Expansion Fund grew from 50 million announced in 2018, but the 2019 round awarded 14.3 million across six schools. We are not going to characterise whether more would have been awarded, because the figure we have comes from a campaigning organisation opposed to selection, and that is advocacy rather than reporting.

And ISEB is reviewing its own examination timetables during the 2026 to 2027 academic year, with 2027 to 2028 dates to follow. So if you are planning two or more years out, the date you write in your diary is provisional.

• Free school meals: all Universal Credit households from September 2026, but families still have to apply.
• The February 2026 white paper mentions selective schools and grammar schools zero times, though it does propose consulting on admissions code changes.
• The prohibition on establishing new grammar schools remains; existing schools may expand only.

Girls' grammar schools

About 61 state grammar schools, entry at 11, and a folklore about English weighting that the evidence does not support.

State grammar schools, England

163

Of which girls-only

about 61

Entry age

usually 11

England has 163 fully selective grammar schools, listed by the 1998 Grammar School Ballots Regulations. Around 61 of them are girls-only, 56 are boys-only, and the rest take both. Those three numbers come from counting that statutory list by hand, so treat them as indicative rather than official: the Department for Education publishes no equivalent breakdown, and the figures circulating on the web mostly trace back to a Wikipedia list.

The 163 figure itself is solid. The National Grammar Schools Association says England has 164, a difference of one that is well within the noise of two sources updating on different days. Northern Ireland has a further 69, which are not part of the English list and follow a different admissions system.

The girls' sector is unusually concentrated. Kent and Medway alone account for 13 of them, and the common pattern is a pair: a town with a boys' grammar and a girls' grammar, the way Tunbridge Wells, Dover, Reading and Plymouth each have one of each. In Kent, every selective school in the Maidstone and Malling planning group is single-sex. That matters when you are checking which papers a sibling sat, because the two schools often share a consortium paper.

Do girls' schools test differently?

In the state sector, no evidence of a systematic difference. In the independent sector, yes, and it is a difference of shape rather than content.

You will hear it said constantly: girls' grammar schools weigh English and verbal reasoning far more heavily than boys' schools do. The number behind it is real. Some consortia weight English and verbal reasoning together at 50%, with maths at 25% and non-verbal reasoning at 25%. But those are consortium papers serving both sexes, and where the same paper is sat by a girls' school and a boys' school the weighting is identical for both. Warwickshire runs one GL paper across six grammars, three of them girls', and all six sit the same thing. Buckinghamshire reports the same 50/25/25 split for its girls' schools as for its boys'.

So the honest answer is that we found no authoritative source showing a systematic gender difference in the state sector, in either direction. The folklore is probably a reasonable generalisation from particular local consortia, read as a national rule.

The difference that is real, and well documented, is in the independent girls' sector. Cheltenham Ladies' College does not run a multiple-choice paper at all. Its assessment day combines group discussions and activities with written and adaptive online tests in maths, English and verbal reasoning, and the decision uses the current head teacher's reference as well as the day itself. Non-verbal reasoning is absent. That is a different shape of process, not a different weighting, and it is the kind of thing a family considering an independent girls' school needs to know several years ahead.

A real girls' school timeline, with dates

What a family with a September 2027 entry has already done and what is still ahead.

If your child is going up for September 2027, the registration deadlines above have mostly passed. That is worth knowing early rather than late, because the window between registration and the test is the one part of this a parent can actually control.

For the independent route the shape is different again. ISEB Common Entrance at 11+ runs two sessions a year, in November and January, with most children sitting the January one, and the papers are English, mathematics and science. It costs 130 pounds, or 50 if your child is moving up from the school's own prep school. ISEB's Common Pre-Tests are a different thing: four adaptive online tests covering the same four subjects in about two hours fifteen minutes, sat once in a window running from 1 September 2026 to late May 2027, and the result is shared, so every school that accepts it sees the same score without your child sitting it twice.

What the schools say about preparation

The published advice from the exam providers and the schools, quoted rather than paraphrased.

This is the part that surprises parents, so it is worth quoting rather than summarising. Cambridge describes its assessments as designed so that children can demonstrate what they are capable of without coaching and excessive preparation. Lancaster Girls' Grammar School, which sets its own FSCE test covering English and maths, says the test is built to remove any need for tutoring, that nothing in it should surprise or intimidate a child, and that no tutor or parent will have knowledge of the test that others lack.

GL draws the same line between two of its own products. Its free familiarisation papers aim for general familiarity with the style of test. Its paid practice papers offer comprehensive preparation. Only one of those is a claim that practice will help you, and it is the one you pay for.

The preparation advice published by a girls' grammar school is worth reading in full, because it is not what most families expect: work well at primary school with maximum attendance, read widely and deeply, keep maths fluent through real day-to-day problems, have imaginative and critical discussions at the dinner table, and focus on calmness and confidence rather than pressure. Notice what is absent. No mention of past papers, of drilling, or of a target number of questions a week.

• Cambridge: assessments let children show their potential without coaching and excessive preparation.
• Lancaster Girls' Grammar: the test is designed so no tutor has an advantage.
• GL: free papers are for familiarity with the format; only the paid papers claim preparation value.
• Free school meals eligibility widened in September 2026 to all Universal Credit households, including those above the income cap. You still have to apply, and it matters because pupil premium status features in some oversubscription criteria.

Mixed grammar schools

The largest and least distinctive group, and the one most families are actually applying to.

Most children who sit the 11+ in England are applying to a co-educational grammar school, and for them the process is the one described in the sections above: one test, one ranked application, 31 October, 1 March. There is nothing different about a mixed school in the process itself, which is worth saying because it is the question families most often ask and the answer is usually duller than they expect.

What varies between mixed schools is the paper, not the system. A mixed school in Warwickshire sits a GL paper shared with five other grammars, three of which are single-sex. A mixed school in Buckinghamshire reports a 50/25/25 split between English and verbal reasoning combined, maths, and non-verbal reasoning. A mixed school in Gloucestershire, from 2028, will sit a curriculum-based test with non-verbal reasoning removed entirely.

The other thing that varies is competition, and it is geographic rather than structural. In a district with one grammar school and a healthy oversubscription, the test matters. In a district with two and a catchment of manageable size, it may not. Thirty-five local authorities have any grammar school at all, so most families in England have no decision to make here.

• The co-ed schools are the ones with the widest spread of entry scores, because they draw from the widest range of catchments.
• A single consortium paper shared across single-sex and co-ed schools means the weighting is identical for all of them, which is worth knowing if you have heard that one sex is weighted differently.

What competing for a co-ed place is actually like

It is geographic, not structural, and the honest national statistics will not tell you.

Most children who sit the 11+ in England are applying to a co-educational grammar school, so this is the section that probably describes your situation. The process is the one set out elsewhere on this page: one test, one ranked application through your local authority, 31 October, offers on 1 March. Nothing about a mixed school is different in the mechanics.

What differs is the range of scores behind the offer. A co-ed school draws from the widest catchment of any of the three types, so the distance between the offer threshold and the school's floor tends to be wider than at a single-sex school in a single-sex town. Practically, that means a mixed school is more likely to be a genuine stretch for a strong candidate and more likely to be a realistic target for a middling one, within the same town.

On how hard it is to get in, the national statistics are close to useless here, and it is worth saying why rather than just declining to quote them. The 83.6% first-choice rate covers every secondary applicant in England, and the overwhelming majority of those applicants are applying to non-selective schools where a place is not in question. The DfE does publish a breakdown by whether an authority has selective schools; we could not retrieve it. So the honest position is that no reliable national figure for grammar school competition exists in the public releases we could reach.

What can be said is geographic, and it is a genuinely useful thing for a family to know. Only 35 local authorities in England have any grammar school at all, and around 60% of them sit in just 11 authorities. So most families are not competing for a grammar place at all, because there is no grammar school to compete for. A few are competing for the most selective thing they will ever do, and in those areas the competition is intense. There is very little in between.

Two local signals are more informative than any national percentage. A catchment area where the grammar draws noticeably fewer disadvantaged pupils than the rest of the area is a catchment where the admissions test is doing less work than the school's results suggest, and Schools Week reported in September 2026 that free school meal rates at the average grammar were 9.2 percentage points lower than elsewhere in their own catchments. And a raw number is more useful than a percentage: in Gloucestershire about 2,500 children a year sit for seven grammar schools, which is the kind of ratio that actually tells a family something about their chances.

• Co-ed schools draw from the widest catchment, so the gap between the offer threshold and the floor is usually wider than at a single-sex school in a single-sex town.
• 35 local authorities have any grammar school; around 60% of them are in 11. Most families are not competing at all, and a few are competing hard.
• A grammar drawing fewer disadvantaged pupils than its own catchment is a place where the admissions test is doing less work than the results suggest.

What happens when a council redesigns the test

Gloucestershire changed everything at once. It is the clearest recent example of what to ask about.

If you are in an area where the test is being reconsidered, Gloucestershire is the worked example, and it is worth reading as a template of the questions to ask your own council.

The seven Gloucestershire grammar schools appointed the not-for-profit Future Stories Community Enterprise to run their Year 7 entrance test for entry from 2028, first sitting in summer 2027. Three things changed at once. The test moved from September of Year 6 to the summer term of Year 5, which is two years earlier and is designed to stop children losing focus over the summer. It became based on the national curriculum rather than reasoning-style questions. And non-verbal reasoning was removed entirely, on the stated grounds that it is the component that typically features in private tutoring.

The reasoning behind the second and third changes is worth noting carefully, because it is a system arguing against itself. The profession identified non-verbal reasoning as the component most dependent on commercial tutoring, and in the same county, most exposed to the commercial tutoring market. Removing it reduces the value of that market to families in the area.

What has happened since is the more useful part. Families interviewed by the BBC were not persuaded. One said they would probably still tutor because everyone else is and you have to try to level the playing field for your own child. Another said it was really expensive and they knew parents who had paid for years and whose children did not get in. A local tutor's assessment was that tutoring will adapt rather than disappear, and that as long as school provision and resources differ, some children will remain better positioned than others.

So if your own council is reviewing its test, the realistic expectation is that the test will get easier to prepare for and the commercial tutoring market will find something else to sell. Plan for the second.

• Gloucestershire's redesign: test moved two years earlier, made curriculum-based, and non-verbal reasoning removed.
• The stated reason for removing non-verbal reasoning is that it is the component that typically features in private tutoring.
• Families in the county were not persuaded that the change would reduce tutoring, and their reasons are worth hearing.

What "super selective" actually means

It is not an official category. It describes schools that select almost entirely at 13, do not roll admissions, and are heavily oversubscribed.

Super selective is not an official term. It has no legal, regulatory or exam-board status; we checked the ISEB site and the admissions pages of Winchester, Westminster and Dulwich, and the phrase appears on none of them. It is descriptive language from the education press, and it is worth defining functionally rather than treating as a category.

The four features the schools have in common are these. Almost all new pupils arrive at one age, usually 13. Selection happens at 13, by entrance test, interview and a report from the current school. They do not accept applications throughout the year, unlike most independent schools. And they are heavily oversubscribed.

The distinction that matters most is from the state system, and it is the age. State grammar schools select at 11. These select at 13. A family that has spent two years preparing for an 11+ is not two-thirds of the way to one of these; they are at the start of a different process.

• Eton takes boys aged 13 to 18 into Year 9, with a sixth-form intake of about 20 places.
• Westminster's Under School takes 4+, 7+ and 11+, and the 11+ place is a through-school place into Year 9 with no further entry testing.
• St Paul's runs a prep school for 7 to 13 and a senior school for 13 to 18.
• Tonbridge is unusual in offering three entry points: Year 9, Year 10 and Year 12.
• Whitgift offers 11+, 13+, 14+ and 16+. Dulwich covers 7 to 18 with entry in Year 7, Year 9 and Year 12.

They do not share an exam

Checking the arrangements one school at a time, because generalising here is how families go wrong.

Three arrangements are worth pulling out, because they are the ones that surprise people.

Winchester does not use ISEB at all. Selection is its own General Ability Testing, school references, a writing assessment and an interview. It then publishes, in plain terms, that the Winchester Entrance papers its candidates sit in April and May of Year 8 are not decisive: entry is not dependent on the results, and the papers are an indicator of academic progress rather than a gate. The school's own academic scholarship exam, called Election, is taken in Year 8.

Dulwich excludes ISEB for one whole route. For deferred entry to Year 9 it states that candidates sit its own tests at the College, and says directly: we do not use the ISEB pre-test for deferred entry applicants.

St Paul's is a hybrid, and a candid one. ISEB does the first filter, St Paul's own papers on a published syllabus do the second, and an interview does the third. It publishes sample questions for its own papers, which is a reasonable thing for a school to do and not many do.

The 11+ is not the 13+, and one test is used for both

The label on a test tells you nothing about the age you take it or what it selects for.

This confuses more families than any other point on this page, and it confuses them for a specific reason: the ISEB Pre-Tests are marketed as the 11+. ISEB itself describes them as sometimes referred to as the 11+, the Pre-Tests, or the ISEB, and as four adaptive online multiple-choice tests used by leading independent senior schools.

But Eton, Westminster, Harrow and St Paul's all use those same Pre-Tests in Year 6 as the first stage of thirteen-plus selection. Eton calls it Stage 1 of assessment for a conditional place. Harrow requires all boys to sit it in the autumn of Year 6. The same paper, sold as a primary-age test, is doing a secondary-age job.

The two processes differ in ways that matter. A thirteen-plus route adds writing and reading analysis that do not exist at 11+; Dulwich's writing test is twenty minutes, five of planning and fifteen of writing. The school report carries real weight, and at Tonbridge the whole route is built around it. The interview is a substantial part of the decision, and there is no 11+ equivalent of it. And at several schools the two are mutually exclusive: St Paul's says it is not possible to sit both the 11+ and 13+ processes and the offer is only for the entry point assessed, and Westminster says those applying for its Under School cannot at the same time apply for 13+ to the senior school.

So if a family has bought 11+ practice papers and mock exams written for nine-year-olds, that is not preparation for any of these. It is preparation for a different exam, at a different age, for a different thing.

• ISEB Common Entrance at 13+ has three compulsory papers, English, maths and science, plus optionals across eight subjects.
• Eton, Westminster, St Paul's and Tonbridge all request a head teacher's report or reference.
• St Paul's runs a second-chance route: candidates who show potential but cannot be offered a place are invited back in Year 8.
• Winchester's 11+ route runs through its feeder school, The Pilgrims' School.
• St Paul's says that for boys from state primaries, the 11+ entry point is the most appropriate, because it allows two years in the junior school.

What it costs, and what help exists

Fees per term, the non-refundable extras, and the bursary programmes in the schools' own figures.

Day fees, lowest

£10,470

Boarding, lowest

£17,322

Boarding, highest

£23,094

VAT on fees

20% since Jan 2025

Fees are reviewed every year, and the movement is visible in the published figures. Whitgift's day fee went from £10,080 in 2025 to £10,470 for 2026. Westminster says the governing body typically reviews next year's fees by the end of the summer term, publishing updates in early July. Dulwich says the governors reserve the right to review fees at any time. So treat any figure here as a snapshot and re-check.

There are also non-refundable charges before a child has started: Eton £480 registration and £3,840 acceptance, Westminster £360, St Paul's £240, Harrow £450 registration plus £2,400 entry and £2,400 final confirmation deposits, Tonbridge £350 registration and £2,250 admissions fee, Whitgift £205 entrance exam fee.

All four headline schools run means-tested bursaries, and all four are explicit that the money is limited. Eton's budget for scholarships and bursaries was £9.7 million in 2024 to 2025, and in the year before that 247 boys received fee reductions, 99 paid nothing at all, and the average award was 71% of the fee, with 18% of boys receiving support. Eton publishes no income thresholds and assesses case by case, and says plainly that it is still not possible to support all applications. Winchester runs a programme of more than £5 million a year and calls it one of the most generous in the country, with awards from 5% to 100% of fees and help with extras such as laptops and trips. Westminster's programme is around £2 million, with means-tested bursaries of up to 100% of day fees. St Paul's is the only one of the four publishing income thresholds: up to £144,000 gross household income, and from 2025 to 2026 a 100% fee award for households below £78,000.

One thing worth knowing about process: at Westminster you must declare an intention to apply for a bursary at the point of registration, and the application itself is only completed if a place is offered. Missing that step is not recoverable.

How competitive, honestly

The published figures are shortlisting figures, not offer rates, and the difference matters.

This is where a lot of published material misleads, so it is worth being precise about what the numbers are.

Westminster publishes that it receives around 500 applications and invites approximately 250 for interview. Dulwich publishes how many it invites: 40 boys for Year 9 entry in September 2027, and 120 for deferred entry in September 2029. Harrow takes around 160 boys into Year 9 and warns that boys registering after 30 September of Year 6 will be competing for fewer places.

Those are invitation-stage figures. They tell you about shortlisting, not about how many offers followed. Westminster's 500 to 250 says nothing about the offer rate at the end, and that number is not published. No school we checked publishes an application-to-offer ratio, and Eton and Winchester do not publish application or offer numbers at all. A figure you will see repeated, that around 400 boys apply to Winchester each year for roughly 140 places, appears only on commercial tutoring websites; we could not verify it against any school or press source, so it is not repeated here as fact.

What can be said plainly is that the funnel keeps going well past the interview. A typical round runs about eighteen months, from registration in Year 5 or 6 through interviews in the spring of Year 8, and several schools require a Year 8 exam after the interview that can still be failed. St Paul's runs a second-chance route in Year 8 for candidates who showed potential but were not offered a place, which tells you something useful about what these schools are looking for: not a single test taken once.

• No school in this tier publishes an application-to-offer ratio.
• Eton's admissions policy is not public; it is available on request or through the parent portal.
• None of these schools publishes a weighting between the written test and the interview. The closest published statement is Dulwich's, which names three components without weights: the examination, an interview, and a reference from the head of the current school.

What these schools say they are looking for

In their own words, and the answer is not practice papers.

The most striking thing about this group is how little they ask for. Westminster says its pre-interview tests are online and that there is no need for special preparation. Winchester says the process should not be stressful or require additional tutoring. ISEB, the exam board that sets the papers, says your child is already excellently prepared because the Pre-Tests are based on the national curriculum, and that additional preparation is not required.

What the interviews reward is described just as plainly. Winchester looks for personal qualities as much as academic potential: curiosity, kindness, integrity, resilience, and a willingness to contribute to the life of a house, conducted by a member of the pastoral care team. St Paul's looks at numeracy, the ability to critically analyse source material, co-curricular interests and general conversation. Tonbridge describes what it wants as talent that can move from asking the right questions to challenging the answers.

Read those three descriptions together and the implication is consistent. Reading, conversation about what a child actually does, and a real account of their interests are what is being assessed. Those are the things a family has either been doing or has not.

One honest tension worth naming: ISEB sells paid preparation materials, including a Pre-Tests prep pack and a parent toolkit, while stating elsewhere that additional preparation is not required. Both are true. The exam board is selling you revision for a test your child is already ready for, and saying so in the same breath.

• Winchester: curiosity, kindness, integrity, resilience, willingness to contribute.
• St Paul's: numeracy, critical analysis of source material, co-curricular interests, general conversation.
• Tonbridge: talent that moves from asking the right questions to challenging the answers.
• Tonbridge's Experience Afternoon is group activities followed by a one-to-one interview, with no written test at first stage.

When to start, and what that means at each age

No authority publishes a year-by-year schedule. This is assembled from what the providers and the evidence actually say.

We should say plainly that no authority publishes a year-by-year 11+ preparation schedule, and we could not find one. What follows is a reasoned default built from the sourced principles, not doctrine. Treat it as a starting position you can adjust for your child.

Years 1 to 4: no 11+ work at all. Reading for pleasure and classroom fluency. This is not timidity, it is what the evidence supports. The EEF finds individualised instruction's effects are actually smaller in primary schools than secondary, and ISEB states the tests follow the national curriculum, so there is nothing to pre-empt in Years 1 to 4.

Year 5, age 9 to 10: learn the format, not the content. Work through one free GL familiarisation booklet per subject at your own pace. This is the point where the timings and the mechanics get rehearsed: a paper, multiple choice, answers in pencil because the sheets are computer-marked, and the instruction pages are not timed, only the numbered questions. GL's own guidance is to familiarise children with the style and format of the test, which is a different and much smaller claim than teaching them the test.

Summer before Year 6: one timed paper per subject. GL says tackling a practice paper under timed conditions gives children valuable experience of managing their time effectively, which is the actual skill being trained.

Year 6: if you buy anything, buy one subject at a time and use it diagnostically rather than as a measure. A practice score should create a picture of your child's strengths and weaknesses, which is a different use from deciding whether they are good enough. Keep the EEF's dosage shape in mind, which is short and regular over a set period rather than a cram, and note its own finding that maths gains much less from tuition than literacy does.

A content ceiling worth knowing: GL says its English and maths papers are based mainly on curriculum content reached during the early part of Year 6, and elsewhere says up to the end of Year 5. Slightly different wording, same meaning. If your child is comfortably past that, the content is not the thing to practise; the format is.

• Years 1 to 4: reading and classroom fluency, nothing test-shaped.
• Year 5: the format, using the free familiarisation papers.
• Summer before Year 6: one timed paper per subject.
• Year 6: one subject at a time, used to find weaknesses rather than to measure worth.

The case against over-preparation

Not a counter-argument for its own sake. It is what the schools, the exam boards and the DfE's own evaluation say.

The strongest evidence here comes from the Department for Education's independent evaluation of the National Tutoring Programme, Year 3, published in September 2024. It found that although many results reached statistical significance, the effect sizes were small and equated to one month's additional progress or less. Two years later, at Key Stage 4, effects equated to less than one month's additional progress. It detected a small longer-term benefit on Key Stage 2 maths only, and no other longer-term benefits at all.

That is a far more modest return than the EEF's five months for well-targeted tuition, and the DfE notes two reasons the figure may even overstate what happened: dilution, where schools bought tuition for more children than could be served, and selection bias, where the children who got it were not a random sample. It says the true impact may be consistent with the expected impact from previous research, which is the fair reading.

The exam boards take the same position from the other direction, because they design against it. GL says its tests are designed to be less influenced by intensive or repeated preparation. Cambridge says it introduces new assessment paradigms to reduce the predictability of assessment content, and sells no practice papers at all. ISEB says adaptive testing means children do not spend time on questions that are too hard or too easy, and states that the Pre-Tests are based on the national curriculum so additional preparation is not a requirement.

On the anxiety question, one school leader said something that is worth more than any general warning. When Gloucestershire moved its test to the summer of Year 5, one of the justifications was that children lose their ability to focus on the test over the summer holidays when they are not at school. That is a head teacher saying that a summer spent drilling is measurably counterproductive. It is also, not coincidentally, the argument for moving the test earlier.

And the thing parents most often report, from the BBC's coverage of the Gloucestershire change: some parents paid for tutoring for years and their children did not get in. That is not an argument that tutoring is useless. It is an argument that the thing being bought is not a guarantee, and the page earlier in this section on competitiveness is where the numbers for that live.

• DfE, National Tutoring Programme Year 3: one month's progress or less at Key Stage 2, less than one month at Key Stage 4, no other longer-term benefit detected.
• The DfE flags dilution and selection bias as reasons the figure may be optimistic.
• GL, Cambridge and ISEB all state in their own words that they design their tests to be resistant to coaching.
• A summer spent drilling is described by a school leader as reducing a child's ability to focus on the test.

What a child can do that is not preparation

The statistics on children's reading are the most alarming numbers on this page, and they are not about the 11+.

Enjoy reading in their free time

32.7%

Read something daily

18.7%

Down since 2005

36% and 20 points

The National Literacy Trust's annual survey, run in early 2025 with nearly 115,000 responses from 5 to 18 year olds, is the current benchmark. It found that 32.7% of 8 to 18 year olds said they enjoyed reading in their free time, which the Trust describes as the lowest in 20 years and a 36% fall since 2005. Only 18.7% read something daily in their free time, again the lowest recorded, down nearly 20 percentage points since 2005. Among the youngest children, those aged 5 to 8, daily reading fell to 44.5%.

Those are the numbers to sit with. A generation of children is reading less for pleasure than at any point in twenty years, in the run-up to the exam that tests comprehension and verbal reasoning. Whatever else a family does, this is the thing that will matter most.

The Trust also reports what pulls children in, which is useful practically: material tied to a favourite film or TV series, at 38.1%, and material matching their own interests, at 37.1%. A child who will not read a book will read something connected to the thing they care about.

On parents of very young children, the Trust's 2026 survey found 51% read with their child daily and 74% chatted daily, and noted that daily engagement remains below 2019 levels across every activity it measures.

One thing we want to be straight about: we could not verify, from a peer-reviewed source read for this page, the size of the link between reading for pleasure and later 11+ scores. That link is widely asserted and intuitively obvious, but it is not quantified here, and this page does not put a number on it. What the survey does show, unambiguously, is that children are reading less.

• 32.7% of 8 to 18 year olds enjoy reading in their free time: the lowest in 20 years, down 36% since 2005.
• 18.7% read something daily in their free time, also the lowest recorded.
• Among 5 to 8 year olds, daily reading fell to 44.5%.
• What works practically: a favourite film or TV series, at 38.1%, and matching the child's own interests, at 37.1%.

Where vocabulary actually fits

Honest about the size of it, because the honest size is still worth doing every day.

We will not tell you vocabulary wins the 11+. It does not, on its own, and anyone who says otherwise is selling something. What the papers do is test comprehension, verbal reasoning, maths and non-verbal reasoning, and a child cannot do any of those well without words. A child who does not know what a word means cannot reason about a sentence that contains it, and verbal reasoning in particular is reasoning with words rather than reading comprehension, which is why GL includes it in most papers.

So the honest claim is narrow. Vocabulary is a floor, not a ceiling. It is the thing that has to be in place before any of the reasoning work is possible, it is the one component a family can work on at home daily without buying anything, and it is cumulative in a way that timed papers under pressure are not. A word learned in Year 4 is still there in Year 6.

What vocabulary practice should not be is a list to memorise the night before. That produces a child who can define a word they have never seen used, which is not what any of these papers asks for. Every paper in the range requires the word in a sentence, in context, to be understood or produced. A child who has only ever met a word on a flashcard has not met the word.

That is also why this app asks for a word from five different directions rather than drilling one. A word that can be defined but never spotted in a sentence is not learned, and the app is built so that it keeps coming back until it is.

• GL: most 11+ papers contain a verbal reasoning test because it is, in GL's words, the best predictor of future academic potential. GL sells that test, so treat the emphasis as interested.
• A word learned in Year 4 is still there in Year 6. Vocabulary is the one component that compounds quietly.
• It is a floor rather than a ceiling, and it is the only major component a family can practise daily at no cost.

What this app does, precisely

A short daily round of vocabulary questions, matched to the child, with a record of what has stuck.

Vocabulary is the part of the 11+ that can be practised honestly. It is also the part most likely to be done badly: a list of words to memorise, drilled the night before, and forgotten by the morning of the test. The reason is that most vocabulary practice has no idea what a child already knows, so it spends its time on words the child would have got anyway.

This app does the opposite. It keeps a record of every word and when it was last seen, and builds each round from words that are new, words that have been slipped, and words that have been got right enough times to count as learned. Words the child already knows stop appearing. Words they nearly knew come back sooner. A word that was got right only because the options gave it away is not counted as known.

Five kinds of question are drawn from the same word: choose the meaning, choose a similar word, choose an opposite word, name the word from its meaning, and fill the gap in a sentence that uses it. Each one asks for the word from a different direction, because a child who can define a word but has never seen it in a sentence has not really learned it.

If the child gets one wrong, the app then says what the option they picked actually means. Choosing debris instead of timid teaches nothing about debris unless the app says what debris is, and a wrong answer is exactly when a child most wants to know the word they did not pick, because they will meet it again.

• Every word has a plain meaning, a sentence that uses it, and real synonyms and antonyms rather than near-misses.
• Difficulty is matched to the child, and wrong options are drawn from the same difficulty band, so an easy question is not lost to an obscure answer.
• A word appears again until it has been recalled on its own a set number of times, spread over days. It is not retired by being seen often.
• Nothing is timed under pressure by default, and a wrong answer never costs a life, a streak or a mark.
• A parent can add their own words, exclude words the child should not see, and print practice sheets on A4.

What this app is not

It is not a test, a tutor, a predictor, or a promise.

It does not predict whether a child will get a place anywhere, and it will not tell you a score means a school is reachable. Anyone who offers you a score-to-school mapping is guessing, and a guess that flatters is a commercial act, not a helpful one.

It does not replace reading. Vocabulary in isolation is not reading, and a child who can define words but does not read will still struggle with the comprehension papers. If your child reads for pleasure, that is doing more for their 11+ than any app, and it is free.

It does not teach the maths, the verbal reasoning or the non-verbal reasoning. It covers one component of the papers, thoroughly.

It does not know your child. It knows what they have answered, which is evidence about their vocabulary and nothing about their ability, their confidence, or how they perform on a day.

Questions parents ask before paying

The ones worth answering plainly.

These are the questions we are asked most, answered as directly as we can. Where the honest answer is that it depends on the school, or that we do not know, that is what is written.

• Is my child ready yet? Readiness is not an age. A child who enjoys reading and can explain a word back to you in their own words is further along than an older child who has been drilled for a term.
• How long should a session be? Long enough to be worth starting, short enough that it happens tomorrow as well. A round that gets skipped is worse than a shorter one that does not.
• What if they get it wrong? They are told what the word they picked means, and the word comes back later. Nothing is taken away and nothing is lost.
• Does this work offline or on a tablet? It is a website, so it works on any modern device. There is no app to install.
• Can more than one child use it? One account is for one child. A second child needs a second account.
• What happens to the data? The account page says what is stored, how long it is kept, and gives a button that deletes the account and everything in it.