Khan Academy for a 10 year old: where it works, where it does not, and the kids it fits
Khan Academy is usually the first thing a parent tries, and for good reason. It is free, there are no ads, nobody is upselling you, and the moment a teacher says your child needs "a bit more practice with fractions" it is the obvious place to send them. It is one of the few genuinely good free things in this category.
It also fails quietly for a specific kind of ten year old, in a specific way, and the failure looks like the child's fault when it is not.
The short answer: Khan Academy works very well when a parent has named one specific gap, sets the topics, and keeps the sessions to about twenty minutes. It works poorly when it is handed to a child as an open instruction to "do some maths", and it needs deliberate mapping if your child is in a British curriculum, IB, or CBSE school, because Khan's default sequence follows US grade levels.
One disclosure before anything else. Khan Academy is a nonprofit, it has no affiliate program, and it pays us nothing. Nothing in this piece earns Cairn a penny either way.
What Khan Academy actually is, mechanically
It helps to be precise about the machine, because most of the good and bad below follows from it.
Content is organised into courses by US grade (4th grade math, 5th grade math, and so on), each course split into units, each unit into skills. A skill has a short video or article and then an exercise. There is a parallel set of "Get ready for" courses that cover only the prerequisites for a grade rather than the full year, which turn out to be the most useful part of the site for a transferring child.
Progress runs through five mastery levels: Not Started, Attempted, Familiar, Proficient, Mastered. A child reaches Familiar by getting 70 to 99 percent of an exercise correct, Proficient at 100 percent, and Mastered only after answering that skill correctly again inside a mixed-skill assessment. Levels can also go down. A child at Proficient who misses a question on that skill later drops back to Familiar.
That last mechanic is worth sitting with, because it is the single most common reason a ten year old walks away. A child who spent Tuesday getting a bar to Proficient can open the site on Saturday and watch it move backwards. For a resilient child that reads as a fair system. For a child who takes correction hard, it reads as punishment, and the site becomes the thing that takes your progress away.
The curriculum problem, in a table
A ten year old is in Year 6 in England, Grade 5 in the US, and Class 5 under CBSE. The year labels differ because England counts school years from age five, which we covered in more detail in the CBSE to British curriculum piece. Same children, three different numbers on the door.
Khan's course numbering is the US one. So a Year 6 parent who types "Year 6" into the site finds nothing, and a parent who reasonably picks "6th grade math" has landed their child a full year ahead. The right default is 5th grade math, or Get ready for 6th grade math if you want prerequisites only.
The bigger issue is what the courses contain. Roughly, for a ten year old:
| Topic your child meets this year | English Year 6 | Where it sits on Khan |
|---|---|---|
| Adding and subtracting fractions, unlike denominators | Yes | 5th grade math |
| Multiplying and dividing fractions | Yes | 5th and 6th grade math |
| Decimal arithmetic, all four operations | Yes | 5th grade math |
| Long division, formal method | Yes | 5th and 6th grade math |
| Percentages | Yes | 6th grade math |
| Ratio and proportion | Yes | 6th grade math |
| Simple algebra, missing number problems, formulae | Yes | 6th grade math |
| Area, perimeter, volume | Yes | 5th grade math |
| Coordinate plane, all four quadrants | Yes | 5th and 6th grade math |
The three rows in bold are the reason a straight "put them in 5th grade math" instruction leaves gaps. Percentages, ratio, and introductory algebra are all in the English Year 6 programme of study and all sit in Khan's 6th grade course. A parent following the US numbering will cover the year's fractions and decimals well and miss three strands the child is being examined on.
The reverse mistake is just as common in the other direction. A CBSE-schooled ten year old is often ahead of Khan's 5th grade computation work and will find most of that course boring, while genuinely needing the decimal notation and data handling units. Putting that child through the full course start to finish is how you turn a strong maths child into one who says maths is boring.
Two caveats, said plainly. Curricula vary by school, and your school's own scheme of work is the final word over any table on the internet. And Khan's course contents do change, so check the unit list on the course page rather than trusting this piece six months from now.
Where Khan Academy is genuinely excellent
Filling one named gap. If you know the strand, Khan is close to the best free tool that exists. Ten minutes of video, then exercises that are well written and that give worked solutions rather than answers when a child asks for a hint. For "she never met decimals as numbers, only as money", this is the fix, and it costs nothing.
Children who like to be left alone. Some ten year olds visibly relax when there is no adult watching them think. Khan lets a child be wrong in private, retry without commentary, and move at their own speed. For a child who freezes when a tutor is waiting on a call, that privacy is the whole product.
A bridge before a new school year. The "Get ready for" courses are built for exactly this: prerequisites only, no filler. A three-week run through Get ready for 6th grade math before September is a better use of a summer than a general tutor, and it is free.
Practice that does not fight you. No ads, no character shop, no streak nagging engineered by a growth team. The gamification is mild by the standards of the category. That matters more than it sounds like it does when the alternative apps are optimised to keep a child on screen.
Where it does not work
As an open instruction. "Do some Khan Academy" is not a plan, and a ten year old given that instruction will do one of two things: drift to skills they have already mastered because the points come easily, or open the course challenge, get most of it wrong, and quietly decide they are bad at maths. Both outcomes look like disengagement and are actually an assignment problem.
For a child who needs a person. Khan has no accountability layer. Nothing notices that your child has not opened it since Tuesday except the parent dashboard, which only tells you what you already suspect. If what your child is missing is someone expecting them at four o'clock on a Wednesday, no amount of free content substitutes for that.
When the reading is the real obstacle. Khan's exercises are word-heavy and written in American English, and for a bilingual child still building academic English the maths is often not what is hard. Watch what happens on a worded question versus a bare calculation. If the gap between the two is large, you have a reading problem wearing a maths costume, and drilling more maths will not touch it.
For a child who takes the level drop badly. See the mastery mechanic above. If your child is the kind who rage-quits a game after one loss, the demotion behaviour is a real risk, and it is worth watching the first week rather than assuming.
The setup that actually works
- Create a parent account at khanacademy.org/parent and add your child from there. Under 13 they cannot hold an independent account anyway, and the parent account becomes their coach automatically. A child can also self-register with your email, which you then have seven days to approve.
- Pick the course by age, not by the year number on their school report. Ten years old means 5th grade math or Get ready for 6th grade math.
- Choose three specific units. Not a course. Three. Write them down where the child can see them.
- Twenty minutes, three or four times a week, at a fixed time. Long enough for a video and two exercises, short enough to finish inside a ten year old's attention.
- Look at the parent dashboard once a week, on the same day. Look at time on task and which skills went backwards, not at points.
- Reassess after four weeks against the three units, not against a general feeling about maths.
On Khanmigo, the paid add-on: it is 4 dollars a month or 44 a year, it adds an AI chat tutor over the same content, and parents can see their child's chat history and get moderation alerts. Whether it earns the four dollars for a ten year old depends almost entirely on whether that child asks questions out loud. Children who narrate their confusion get a lot from it. Children who go silent when stuck get very little, because it waits to be asked. The free site is not a stripped-down version, so start there and add it later if you find your child wanting to talk to it.
Frequently asked
Is Khan Academy really free? Yes. It is a nonprofit, the content and exercises are free with no ads, and only Khanmigo is paid.
Is Khan Academy enough on its own? For closing a specific gap, often yes. As a child's whole maths provision alongside school, usually not, because it does not cover the written reasoning formats British and IB schools examine, and it provides no accountability.
What grade should I put my 10 year old in? 5th grade math, or Get ready for 6th grade math for prerequisites only. Not 6th grade, even if your child is in Year 6.
Does Khan Academy follow the British curriculum? No. Content is sequenced to US grade levels. Most of the mathematics overlaps, but percentages, ratio, and early algebra sit a year later than the English programme of study places them.
How long should a 10 year old spend on it? Twenty minutes a session, three or four times a week, is more effective than an hour on Sunday. Attention, not total minutes, is the constraint at this age.
Is Khanmigo worth it? Only if your child asks questions when stuck. Try the free site for a month first.
The part that is not about Khan Academy
Everything above is really one question wearing a costume: does this tool fit this child. Khan Academy is free, which makes it feel like a decision with no cost, and that is the trap. The cost is six weeks of a ten year old's willingness, which is not free at all, and which you do not get back if the answer turns out to be no.
The questions worth asking before you start are the same for any tool. How long can this child stay with something. What happens when they get it wrong. Do they do better alone or with someone waiting. Is the thing I want fixed actually a maths problem.
That is the problem Cairn was built around: a short conversation about how your child learns, then a weekly plan of one to three specific resources matched to the gaps that are real, with the reasoning for every pick written out. Free resources like this one sit in the plan alongside paid ones, and commission never decides the ranking. You can see how the engine works if you want the mechanics.
And if you would rather have the thinking without the product, the letter below takes one question like this each week.