Skip to content

IXL review: relentless practice that works, if you can live with SmartScore

A vast K–12 skills library with genuinely useful diagnostics — and a scoring mechanic that can reduce a tired child to tears. Excellent for schools, harder to love at the kitchen table.

3.6/5By Dr Amara OseiRe-tested 3 min read
Best forschools and trustsfilling specific gapsexam practiceages 7–16
Abstract illustration in warm oranges representing a grid of practice questions
IXL, tested across two terms with a Year 7 set and two families at home.

Pros and cons

What we liked

  • Enormous coverage: maths, English, science, social studies and Spanish across K–12
  • The Diagnostic pinpoints a working level per strand rather than one blunt grade
  • Analytics are the best we have used for finding the exact skill that collapsed
  • Skill plans map onto most national and state curricula, which saves hours of planning

What we didn’t

  • SmartScore drops steeply for a late wrong answer, which can wipe out twenty minutes of work
  • Explanations appear after a mistake, so it teaches by correction rather than instruction
  • Presentation is functional and dry — expect resistance from younger children
  • Family pricing climbs quickly once you want more than one subject or child

IXL is the most comprehensive practice platform in school education, and one of the most divisive. Ask a head of maths and you will hear about the analytics. Ask a parent and you will hear about SmartScore. Both are right, and the gap between them is the whole review.

We used it for two terms with a mixed-ability Year 7 maths set of twenty-nine, and followed two families using family subscriptions at home with children aged eight and thirteen.

The scale is the point

IXL's skill library is enormous. Maths runs from counting objects in pre-K to calculus, English language alongside it, plus science, social studies and Spanish. Each skill is narrow enough to be useful — not "fractions" but "add fractions with unlike denominators using models" — and each generates fresh questions rather than serving a fixed set, so a child cannot simply memorise their way through.

For a teacher, that granularity is the product. When a class assessment shows a wobble, you can find the four skills that actually address it and set exactly those, rather than assigning a chapter and hoping.

The Diagnostic is the best reason to buy it

IXL's adaptive Diagnostic asks a stream of questions across strands and returns a working level for each one — number sense, algebra, geometry, measurement, data — instead of a single grade. Run it in the first week of term and you get a map of a class that would otherwise take a month of teaching to build.

In our Year 7 set it flagged two children whose arithmetic was two years behind their algebra, which is not a pattern our own baseline test had picked up at all. That single insight changed how we grouped the class for a term.

SmartScore, and why it makes people angry

Every skill has a SmartScore out of 100. Correct answers raise it, wrong answers lower it, and the penalty grows as the score rises. The intent is defensible: it stops a child limping to "mastery" on a coin-flip success rate, and it rewards sustained accuracy rather than volume.

The lived experience is different. A child at 94 who makes two careless mistakes because they are tired can drop into the seventies and lose the sense that the last half hour counted for anything. Both families we followed described the same scene, and both had at some point sat next to a crying child. This is not an edge case — it is the most consistent piece of feedback about IXL anywhere, and IXL has not meaningfully changed it.

Teaching versus testing

Every skill has a "Learn with an example" panel and, for most, a short video. They are clear and mathematically sound. But they sit behind the question, framed as help once you are stuck, and children under about ten rarely go looking for them unprompted.

That makes IXL excellent practice and mediocre instruction. In a school where a teacher has already taught the method, it is close to ideal. At home, where a parent is often the only explanation available, it can turn into forty minutes of a child guessing and an adult improvising a lesson on adding unlike denominators.

The analytics

This is where IXL is genuinely without equal. You can see, per child and per skill, how long they spent, how many questions they answered, where the accuracy fell away, and which specific question types keep going wrong. Trouble Spots surfaces exactly what to reteach.

It is also, in fairness, a lot of data. Teachers new to it should pick two reports and ignore the rest for the first half-term.

What we would use instead

  • For fluency at primary level with far less friction, a dedicated mental maths tool is kinder and cheaper.
  • For instruction first and practice second, a video-led platform with structured courses fits home learning better.
  • For exam-board specific revision at GCSE and A-level, a question bank aligned to the specification will beat a general skills library.

None of those replace IXL's breadth or its Diagnostic. If you need one practice engine across a whole school, it remains the strongest option — as long as somebody has a conversation with staff and parents about SmartScore before it lands.

What it looks like

  • Abstract representation of an IXL skill practice screen, with a question area and a progress meter
  • Abstract representation of an IXL teacher analytics report, with rows of per-skill results

Questions people ask

What exactly is SmartScore and why do people complain about it?
SmartScore is IXL's per-skill progress number, running from 0 to 100, where 80 counts as proficient and 100 as mastery. It rises with correct answers and falls with incorrect ones — and crucially, the drop for a wrong answer grows as your score gets higher. A child who has ground their way to 92 and then makes two careless slips can find themselves back in the seventies. Pedagogically there is a defensible argument for it. In a real kitchen at half past six in the evening, it is the single most common cause of tears we heard about.
Is IXL better for schools or for families?
Schools, clearly. The analytics, skill plans and class management are built for a teacher with thirty children and a scheme of work, and at site-licence prices the value is strong. At home you are paying a premium for a practice engine with limited instruction, and you become the teacher who has to explain the method when your child gets stuck.
Does IXL actually teach, or only test?
Mostly it tests. There is a "Learn with an example" panel and video tutorials attached to most skills, and they are competent, but they are positioned as help after a failure rather than teaching before an attempt. Used as practice on top of real instruction it is excellent. Used as the instruction itself it is thin.
Which subject is IXL strongest at?
Maths, by a distance — the skill tree is the most granular and the question generation is the most robust. English language arts is solid on grammar and mechanics and weaker on anything requiring extended writing. Science and social studies feel like later additions and cover less ground.

About the reviewer

Reviewed by

Dr Amara Osei

Senior reviewer — maths & science

Former head of maths turned edtech reviewer. Amara has spent eleven years teaching secondary maths and four evaluating software for a multi-academy trust.

  • PhD Mathematics Education
  • PGCE Secondary Mathematics
  • 11 years classroom teaching

3 comments

Parents, teachers and learners — tell us how this went for you.

  1. Katherine Ellis

    Thank you for saying the SmartScore thing out loud. Our school introduced IXL with no warning about it and I had a very unhappy nine year old for about three weeks. The advice to set a time limit instead of a score target has genuinely helped.

  2. Mr Adeyemi

    Head of maths here. The Diagnostic really is the killer feature and it is why we renewed. My only addition is that you need to ringfence time for staff training on the reports, otherwise everyone just uses it to set homework and you have paid for a very expensive worksheet generator.

  3. Jo

    Agree with the point about instruction. My son is thirteen and perfectly capable of reading the worked example, and it works well for him. My eight year old needs a person sitting there, which rather defeats the object of buying software.

Add your experience

Tell us how this worked for your learner: their age, what stuck, what didn't. Every comment is read by an editor before it goes live, and links aren't allowed.

Shown with your comment

Never published — used only if we need to reply

25–800 characters, no links

25 more characters needed

Preparing the form…