IXL Review – It Works, and Children Cry. Both Are True.

The Strongest Evidence in This Review Series

We have spent this series asking one question over and over — what is the actual evidence? — and mostly being disappointed. Testimonials standing in for trials. Company-funded “case studies”. Testing described but never published.

Independent, randomised, controlled, and assessed against a government-defined standard. That is a genuinely strong position and we are going to give it full credit — it is more than almost anything else we have reviewed can claim.

There is a second study too: SRI International validated an analysis of 179 Oklahoma public schools as meeting ESSA Tier 2 and What Works Clearinghouse standards.

One claim we would discount, in fairness to the rest: the marketing line that schools using IXL rank “as much as 15 percentile points higher” on state assessments. That is correlational and self-selecting — schools that buy and successfully implement educational software differ in funding, leadership and staffing from schools that do not. The RCT is the finding that matters; the percentile claim adds nothing to it.

Education Codified the Ladder We Built by Hand

Something worth pausing on, because it reflects on every other review on this site.

Across this series we have applied an informal evidence hierarchy — randomised controlled trials above observational studies, independent testing above company claims, published data above described process. US education law has that hierarchy written into statute.

The supplement industry has no equivalent, and the difference is stark. A school district can ask “which tier?” and get a checkable answer. A person buying a CBD tincture or a weight-loss supplement has no such vocabulary, which is why we have spent this series building one by hand — asking whether a trial was randomised, whether the assay was independent, whether a certificate was published rather than promised.

It is also why a Tier 1 claim is worth something. There is a defined standard, an external body applying it, and a consequence for meeting it — which is exactly what “lab tested” on a supplement label lacks.

The SmartScore, and Why It Hurts

Now the other half, because the consumer picture is close to the opposite of the research picture, and almost all of it traces to one design decision.

Consider how that feels from the inside. A child works to 92, makes one error, lands back in the 70s, and now has to correctly answer a long run of questions they have already demonstrated they can do — not to learn anything, but to get back to where they were.

The pedagogical logic is defensible, and we should say so: mastery should mean consistent accuracy, not a lucky streak, and a metric that cannot fall is not measuring mastery. The problem is not that it exists. It is the magnitude and the timing — the penalty is largest exactly when the child is closest to finishing and most invested in the outcome.

Reported consequences include anxiety, stress, frustration, anger and disengagement, with descriptions of meltdowns and physical symptoms such as stomachaches. The most incisive criticism in what we read is that it ties self-worth to a score rather than to comprehension — the child stops asking “do I understand this?” and starts asking “why won’t the number go up?”

Below 2 Out of 5, and Children in Tears

We treat consumer ratings carefully on this site — unhappy customers write more reviews than happy ones, and a low score alone proves little. This case is unusual enough to report prominently.

• Trustpilot and Sitejabber both sit below 2 out of 5, with reviews repeatedly describing children in tears.
• Common Sense Media parents describe it as overwhelming and frustrating for children who struggle.
• There is a Change.org petition specifically calling for the SmartScore system to be eliminated.

A fair caveat that cuts the other way: most IXL use is assigned by schools, and an unwilling user is not a customer who chose the product. Some share of the anger is really anger at compulsory homework, arriving at the nearest available target. We do not think that explains the specificity of the SmartScore complaint — but it plausibly explains some of the volume.

How Both Things Are True at Once

A Tier 1 RCT and a sub-2/5 rating look contradictory. They are not, and understanding why is the most useful thing on this page.

A demanding drill system can raise measured attainment over a term while making a child dislike the subject. Those are not competing claims about the same thing — they are claims about different things, and the study design that establishes the first is not built to detect the second.

Which matters more depends on the time horizon you care about. If the goal is a better score this year, the RCT is the relevant evidence and it is good. If the goal is a child who does not decide at nine years old that they are bad at maths, the reviews are the relevant evidence and they are alarming. We are not able to tell you which effect is larger over a decade, because nobody has run that study.

What This Teaches About Any Progress Metric

A word on why a site mostly about weight is interested in a maths platform, because the connection is real rather than manufactured.

The SmartScore is a case study in a progress metric designed to motivate that ends up demotivating, and anyone who has weighed themselves daily during a period of weight loss will recognise the mechanism precisely.

The generalisable lesson is the same in both cases: watch the trend, not the punishment. We made this argument reviewing an automatic weight-tracking device — a single reading is noise, the direction across weeks is signal. A metric that can move sharply against you for a single event should be read over time or not at all.

For a parent, that translates into something practical: talk about whether the child understood the topic, not about what the SmartScore says. The number is an instrument, and instruments that make people quit are badly calibrated regardless of what they are measuring.

The Children It Is Worst For

Reporting concentrates the harm in three groups, and the mechanism is plausible in each.

• Perfectionists. A metric that punishes a single error hardest near the finish line is close to worst-case design for a child who already cannot tolerate mistakes.
• Children with ADHD. Banking progress requires sustained error-free concentration across a long run of questions. The design penalises exactly the thing that is hardest, and a momentary lapse costs disproportionately.
• Children with existing maths anxiety. This is the group we would be most cautious about, because the intervention risks reinforcing the belief it is meant to dismantle — that maths is a thing you fail at.

If your child is in none of those groups and responds well to a target, the picture flips entirely — the same mechanic that produces meltdowns in one child produces focused persistence in another. This is a product with unusually high variance between users, which is why the trial and the reviews can both be honest.

Price, and the Guarantee You Lose in the App Store

The core package is good value for what it covers — maths from Pre-K to Calculus, language arts Pre-K to 12, science through physics, and social studies. Twenty dollars a month for a curriculum with that range is not expensive, particularly for homeschooling.

Buy direct. This is now the third distinct product category in which we have found that purchasing through an intermediary silently costs you the maker’s guarantee — and here it is the difference between a genuine 30-day trial and none at all. Given that whether this suits your particular child is the entire question, that trial is the most valuable term on offer.

After 30 days there is no refund, full or partial. Cancelling stops future billing and you keep access to the end of the paid period — so use the 30 days as an actual trial, and decide inside it.

How It Compares

The honest framing is that you are paying for evidence and coverage, and accepting a scoring mechanic that a substantial number of families find punishing. Free alternatives exist and cost nothing to try — trying one first is a reasonable move, particularly if your child is anxious about the subject.

Pros and Cons

How to Use It Without the Damage

Buy directly from IXL.com and treat the 30 days as a genuine trial. Whether this suits your particular child is the whole question, and it is answerable in a fortnight of watching them use it. Buying through an app store forfeits that.

Watch the first sessions rather than the reports. The thing to look for is not the SmartScore — it is your child’s face when the SmartScore drops. That single observation tells you more than any review, including this one.

Set time targets, not score targets. “Fifteen minutes” is achievable and ends on schedule. “Get to 100” can extend indefinitely through no fault of the child, which is the mechanism behind most of the distress reported.

Talk about understanding, not the number. Ask what the topic was and whether it made sense. If the number is the conversation, the tool has redefined what learning means in your house.

Be readier to stop than you expect to be. The evidence says it raises scores over a term; nothing establishes that it is worth a child concluding they are bad at maths. If you see distress, the trade is not worth it — and free alternatives cost nothing to try.

Limitations of This Review

We have not used IXL, and we have no children to observe using it. We have read the trial through press coverage and the vendor’s own summary rather than the full report, so we cannot speak to sample size, attrition, effect size in standard units, or how the control condition was defined — and “10 points on Star Math” is not a figure we can interpret in context. A single-district trial in grades 3 to 5 over one term is also a narrower basis than a Tier 1 label might suggest to a casual reader.

The consumer criticism is aggregated from review platforms and secondary reporting rather than a survey. Review platforms over-represent dissatisfaction, and much IXL use is assigned by schools rather than chosen, so some anger is plausibly aimed at compulsory homework generally. The claims about which children are worst affected are reported patterns and plausible mechanisms, not clinical findings, and we have not seen research isolating harm by student profile.

Pricing and refund terms are as reported and promotional rates change — verify before subscribing, particularly the direct-purchase condition on the guarantee. Nothing here is educational or psychological advice, and persistent distress about schoolwork warrants a conversation with a teacher or clinician rather than a review site.

Frequently Asked Questions

Verdict

Give the evidence its full weight first, because it is genuinely rare. An independent randomised controlled trial run by Johns Hopkins’ Center for Research and Reform in Education, in which IXL Math students significantly outperformed controls by around 10 points on Star Math, meeting ESSA Tier 1 — the top statutory evidence tier — and What Works Clearinghouse standards, with a second study validated by SRI International at Tier 2. Nothing else in this review series can claim that.

It is worth noticing what education has that our usual subjects do not: a four-tier evidence ladder written into law, with an external body applying it. We have spent this series constructing that vocabulary by hand for supplements because no such standard exists there.

Then the SmartScore. It is non-linear: points come easily early, but near mastery a single wrong answer can strip many points, sending a child back to re-answer questions they had already got right. Trustpilot and Sitejabber both sit below 2 out of 5, reviews repeatedly describe children in tears, and there is a petition to remove this specific feature — which is what separates a coherent design complaint from ordinary grumbling.

These do not contradict each other. The trial measured test scores over a term; the reviews measure what it is like to use for years. A demanding drill system can raise attainment and make a child dislike the subject at the same time, and the study design that proves the first is not built to detect the second. Nobody has run the study that would settle it.

The wider lesson is one this site keeps arriving at from different directions: a progress metric that moves sharply against you at the moment you are most invested stops being motivation and becomes punishment. Anyone who has weighed themselves daily and watched the number rise after a good week knows the feeling exactly. Watch the trend, not the punishment — and with a child, talk about whether they understood it, not about the score.

Buy direct from IXL.com, never through an app store, because the 30-day full refund exists only on direct purchases and whether this suits your particular child is the entire question. Set time targets rather than score targets, watch your child’s face when the number drops, and be readier to stop than the evidence alone would suggest.