Ask a seventeen-year-old what they know and they will hand you a transcript. Four years of course titles and letters. It tells you which rooms they sat in and roughly how compliant they were in each one. It does not tell you what they can do, and it is useless the moment they walk into a room where nobody cares about the letters.

In the Five E’s of Education I called the fourth one Expertise, and defined it in three words: prove what you know. This essay is about what that looks like as a practice, and as a product, for someone who is young enough that the practice will compound for sixty years.

Expertise is a ledger, not a feeling

Most people think of expertise as a state you are in. You are an expert in something, or you are not. That framing is useless for a learner, because it gives you nothing to do on Tuesday.

A more useful framing is that expertise is a ledger: a list of specific claims, each of the form “I can do X,” each backed by a piece of evidence, each with a date on which it was last checked. “I can write a paragraph that holds up.” “I can find the break-even price for a product with fixed and variable costs.” “I can get a stranger to agree to a fifteen-minute call.” Small, checkable, and yours.

The ledger does three things a transcript cannot. It makes knowledge specific, so you can see the gaps. It makes knowledge testable, so you can tell the difference between what you learned and what you merely heard. And it makes knowledge legible to strangers, which is the whole point of having it.

The two techniques that actually work

There is a large research literature on how people learn, and most of it is ignored by the people it is about. In 2013 Dunlosky and colleagues reviewed ten common study techniques and rated them by how much evidence supported them (Improving Students’ Learning With Effective Learning Techniques). Two came out on top, and neither is what school teaches you to do.

The first is practice testing: retrieving something from memory, rather than re-reading it. Roediger and Karpicke’s experiments showed that students who studied a passage once and then tested themselves on it remembered far more a week later than students who re-read it three times, even though the re-readers felt more confident (Test-Enhanced Learning). Retrieval is not a way of checking learning. It is the learning.

The second is distributed practice: spacing the retrieval out over days and weeks instead of cramming it into one night. Combined, the two say something simple. Knowledge that is never re-tested decays, and knowledge that is re-tested on a schedule sticks.

Highlighting, re-reading, and summarizing, the three things students spend most of their study time on, all rated low.

What this means for the ledger

Put the research and the ledger together and you get the design of the thing I would build.

Every entry in the ledger is a claim, and every claim is also a future question. “I can find the break-even price” becomes, three weeks later, a problem with new numbers, unannounced. Clear it and the claim stays green and the next check moves further out. Miss it and the claim goes amber, the check comes back sooner, and the ledger tells you exactly which thing you thought you knew and did not. Mastery is not a box you ticked. It is a state you have to keep earning, and the system should make the decay visible rather than pretending it does not happen.

Every entry also has evidence attached, and here is where a teenager’s ledger stops being a study tool and becomes something else. The evidence for “I can write a paragraph that holds up” is a paragraph, published, with a date. The evidence for “I can get a stranger to take a call” is the call. The evidence for “I can build a landing page” is the page. Over four years that pile of evidence is a portfolio, and it was built as a side effect of learning rather than assembled in a panic the month before applications.

The notes go in too, but the notes are not the point. Sönke Ahrens made the case in How to Take Smart Notes that a note is only worth keeping if it is written for your future self and connected to something you already know. In the ledger, a note is worth keeping if it can become a question. If you cannot turn it into something you could be tested on, you have not understood it yet.

The case for portfolios, honestly stated

The standard argument for all this is that employers now care about skills, not credentials, so the kid with a portfolio beats the kid with a transcript. I have made that argument on this site and I want to state it more carefully.

The direction is real. Large employers have removed degree requirements from many postings. But a 2024 study by the Burning Glass Institute and Harvard Business School found that most of the companies that dropped the requirement had not meaningfully changed who they actually hire (Skills-Based Hiring: The Long Road from Rhetoric to Reality). The announcement ran ahead of the behavior, by years.

The lesson is not that portfolios do not matter. It is that proof has to be legible to a stranger in under a minute, or it does not count. A folder of notes is not proof. A ledger of dated claims, each with a link to the evidence and a record of the last time it was checked, is the closest thing to proof a seventeen-year-old can carry into a room. It also happens to be the thing an employer, an investor, or a customer would build for themselves if they had the time.

What I would build, in one paragraph

A single place where a student writes down what they can do, in small claims. Each claim links to evidence and turns into a scheduled retrieval question. Green claims drift out to longer intervals; missed claims come back sooner and show up on the morning screen as the thing to fix first. The public view is the evidence, ordered by what was most recently re-proven, so a stranger sees not what the student once did but what they can still do today. No feed. No followers. A ledger, a schedule, and a pile of proof.

Build it for one kid and it is a study system. Build it for a school and it is the transcript that should have existed all along.

Related: Education: What You Know, Why Schools Fail to Teach the Most Important Skill: Learning How to Learn, How to Become an Exponential Learner, Skills Over Credentials.