Two people score 72%. One is a month from mastery, the other is a year.

A test score tells you how someone did on a set of questions. It does not tell you the thing you actually want to know, which is how far they still have to go and what specifically is in the way.

That gap is the reason we stopped reporting scores and built something else. We call it the Competence Index, and this post is about what it measures and why it turns out to matter more than a mark.

The problem with a number that describes the past

Here is a small piece of maths that makes the point better than any education argument.

Take the number 27. If it's even, halve it. If it's odd, triple it and add one. Repeat. Simple enough to explain to a child and every number you try eventually falls to 1.

But how long it takes is wild. 26 gets to 1 in ten steps. 28 takes eighteen. And 27 climbs all the way up to 9,232 before it comes down, taking 111 steps to get there. Three numbers sitting right next to each other, three completely different journeys. Nothing on the surface of the number tells you the length of the trip.

Test scores work exactly the same way. The score is the number on the front. The thing you needed was the length of the journey still ahead.

Learner A

72%

4 sessions outKnows the material, connected it well, one shaky area.

Learner B

72%

40 sessions out Memorized the right answers on top of two beliefs that are wrong.

Same paper, same mark, very different situations. A score can't separate these two. That's the whole problem.

What we measure instead

Rather than counting what a learner got right, we estimate how much mental work is left before what they know settles into something coherent a picture where the pieces connect and hold up under pressure. We call that quantity Understanding Energy. High energy means a long way to go. It falls as understanding consolidates.

To work it out, we read four things out of an ordinary written answer:

• How much ground they hold. The breadth of what they've actually got.
• How joined-up it is. Isolated facts collapse the moment a question is phrased differently. Connected ideas don't.
• How much of it is wrong. Not gaps  confident, load-bearing beliefs that happen to be false. These are the expensive ones.
• How close it is to clicking. Whether they're near the moment where ten remembered facts compress into one principle and the subject suddenly gets easier.

Those four move independently, which is the point. Someone can learn more and get further from understanding, because they've piled new material on top of a misconception. A percentage can't see that. This can.

What you actually get back

After an assessment, instead of a mark, you see:

• Where you are - a single 0–100 index, plus separate readings for how well you understand it, how connected your knowledge is, and whether you could use it in a real situation.
• How far you have to go - in practice sessions, practice minutes and reading hours. Real units you can put in a week.
• What to do next - the specific topics that will move you the most, ranked by how much each is worth. Not syllabus order, and not simply a list of what you got wrong.
• Whether you're kidding yourself - the gap between what you think you know and what you demonstrably know. Being overconfident is a different problem from not knowing, and it needs a different fix.
• Which way you're heading - accelerating, steady, plateaued or slipping, based on every attempt so far.

Sometimes the number gets worse, and that's good

When you discover that something you were sure about is wrong, your Understanding Energy goes up. You know more than you did an hour ago, and the model says you're further away.

That isn't a bug. It's the most valuable week of the whole process, and it's the week most learners are told they're falling behind. Back to 27: the only way it reaches 1 is by climbing to 9,232 first. Real learning has the same shape - things get messier just before they get clearer.

Because we measure distance rather than performance, we can tell that climb apart from an actual plateau. One means keep going, you're closer than it feels. The other means change the plan. Telling a learner the wrong one is how good people quit.

Why this matters more than a better mark

Once you can measure the distance to a goal in units of effort, you can do something no score allows: say what it will take to close it. How many sessions. How much time with a mentor. Roughly how many weeks.

That turns a report card into a plan. You tell us what you're trying to become, we tell you honestly how far away it is and what's holding you there, and then we give you the mentors and the tools to cross it.

That's the whole idea. The number just had to be worth trusting first.