About us
Markaroo is the adaptive assessment and learning platform that takes students to top grades.
“Mark” plus kangaroo: marking that helps students leap a grade.
Who we are
Most revision tools are built around a simple assumption. Do enough past papers and you will get the grade. Most teachers sense, somewhere, that this is not quite right.
The hard part is not doing the papers. It is knowing why an answer that felt right scored 18 out of 25, which habit is costing the marks every time, and where to point the next hour of revision. Markaroo was built around that reality.
We are a team of teachers and a researcher with a PhD in AI. We spent years in classrooms, marking the papers and writing the same comment for the thirtieth time, before building Markaroo. We knew what the tools were missing and had the means to fix it.
Neither half of that team works alone. AI without pedagogy produces confident feedback a teacher would not sign off. Pedagogy without AI stays stuck at the number of essays one person can read in a week. What a student gets from a good tutor is not more content. It is someone who reads the answer properly and says what to change. That should not depend on what a family can pay.
What we want to achieve
Remove the need for private tuition.
Not supplement it, replace what it does. Tuition sells two things: work marked as accurately as an experienced teacher would mark it, and feedback specific enough to change what the student writes next time. Markaroo delivers both at a price a family or a school can absorb, so a student in a comprehensive class of thirty gets the same feedback loop as a student whose parents can pay for it.
That is the test we hold ourselves to: marking accurate enough that a teacher trusts it, and the Coach’s feedback moving the grade. If those two hold, tuition becomes optional.
What we’re passionate about
Accuracy and consistency
We want to be held to a high standard, and we publish the numbers rather than assert them. Marking is a judgement, and two experienced examiners routinely differ by two to three marks on the same essay. Across our benchmark Markaroo sat 2.8 marks from the awarded mark, ahead of the leading competitor, and it does not drift on the thirtieth script the way a tired marker does at midnight.
Consistency at that level is not a nice to have. It is the difference between feedback a student can act on and feedback that depends on who picked up the pen.
Pupil progress
Feedback is only worth writing if it changes the next answer. Ours is shaped by pedagogical theory rather than by what a language model finds easy to say: name the specific habit costing the marks, ask the counter-question instead of handing over the answer, show the rewrite, then re-mark it so the student sees the gain. The point is genuine understanding of a new idea, and the grade that follows from it.
Lifting the marking burden off teachers
Marking and grading of papers and homework is the load that pushes good teachers out of the profession. Accurate marking with detailed feedback hands that first pass back, and returns 125 hours a year to every teacher, around 1,000 across a department of eight, so teachers spend their time teaching and planning instead of writing the same comment for the thirtieth time.
We want teacher burnout to be a thing of the past, and we want the same tool that relieves the workload to be the one that shows the department where the class is struggling.
The problem
Most students don’t have a knowledge gap. They have an application gap.
They can learn the theory. They can recite the model. Then the exam asks them to apply it to an unfamiliar context, they write a solid answer, score 18 out of 25, and can’t see why it wasn’t more. Most e-learning stops exactly where the marks are won: videos and quizzes that check recall, not application.
Every student is told to do past papers, and most of them do. Then they have to mark their own work. For long-form essay questions, in Economics, History, English, Business, Psychology and the rest, self-marking barely works. A mark scheme is written for a trained examiner, not for the student who wrote the answer. So the practice happens and the improvement doesn’t.
The obvious fix is a teacher with time to read every essay. In a comprehensive class of thirty, that time does not exist, and the teacher is not the problem: the ratio is. In a smaller private setting the same teacher can turn essays around weekly, and the gap that opens up is not ability, it is feedback. Families who can afford it close it with private tuition. Everyone else practises in the dark.
That is the gap Markaroo closes. Examiner-level feedback on real essays, plus practice that trains application rather than recall: in Economics, questions inspired from current Financial Times articles, so students apply the theory to the world the examiner is actually writing about.