With comparative judgment, you read two pieces of writing and decide which is better.

You don't just make one decision, you make a series of decisions, and it isn't just you making the decisions. In our big national projects, thousands of teachers make hundreds of thousands of decisions.

We combine together all of the decisions to create a measurement scale for every piece of writing. Independent research has shown this process to be as reliable as double marking, but much quicker.

Human Comparative Judgement is the gold standard of human decision-making. It is supported by an extensive research literature.

We've now enhanced Comparative Judgement even further with the use of AI. As well as teacher judges. You can now add in AI judges to an assessment.

You can ask the AI judges to complete whatever proportion of your judgments you want. Our recommended approach is to get the AI to do 90% of your judgments. That leaves 10% for humans to do, which is enough for every essay to be seen twice by humans and for the AI model to be validated by all the human decisions.

Let's assume that it takes you 2 hours to mark 30 essays the traditional way with a rubric. It will then take you just half the time. That's 1 hour to mark them using human only Comparative Judgement. If you add in AI judges, it will take you just 6 minutes.

AI enhanced Comparative Judgement provides you with even quicker results, more accurate data and more detailed feedback.

We also provide incredibly extensive and in-depth PDF feedback reports for every pupil. Every pupil gets a personalised feedback report with specific pupil friendly action steps. Teachers get a separate report with more data and extra details.