The content isn’t the problem. The container is.

Every LMS on the market is built around the same noun. You buy courses, assign courses, complete courses, report on courses. SCORM packages a course. Pricing is per seat per course. The dashboard counts courses. The word is so universal that nobody notices it was a design decision.

It was the wrong one, and the evidence has been sitting in the psychology literature since before anyone had a dashboard.

Human memory does not work in courses. It works in intervals. Whether a front-desk agent remembers the chargeback procedure in March is not decided by how good the module was in January. It is decided by whether anything happened in February. A course, by definition, ends. Memory’s only question is what comes after the end, and a course-based system’s answer, by construction, is nothing.

The course’s own exit exam can’t see the problem

In 2006 Roediger and Karpicke ran an experiment L&D should find uncomfortable. Learners either re-read a passage or were tested on it. Five minutes later, re-reading won, 83% to 71%. One week later it had reversed: the tested group held 61%, the re-readers 40%.

Now look at where a course puts its measurement. The end-of-module quiz sits at the five-minute mark, with the content still on the screen behind the learner. It is the one point on the curve where passive delivery looks best, and it is the only point where the course can measure, because once the course ends there is no course left to do the measuring.

The 88% average on your dashboard isn’t wrong. It’s a receipt. It confirms the content was delivered. It says nothing about whether any of it was kept, and it was never in a position to.

The schedule is the intervention

Cepeda, Pashler, Vul, Wixted and Rohrer reviewed 317 experiments on spaced versus massed study and found the expected advantage for spacing, roughly 15% on average. The finding that matters more is the one that rarely gets quoted: the optimal gap between revisits grows with how long you need the memory to last. Something that must be remembered in six months wants revisits weeks apart. Not a longer course. Not a better course. Gaps.

Which means the thing that actually produces retention lives between sessions, in the empty space that a course-shaped system doesn’t model, can’t schedule and doesn’t report on. You can build the best module in the industry, and the thing that would have made it stick is outside the object you built.

Kerfoot’s randomised trials with 537 residents show what that looks like drawn as two lines. Identical content, delivered once as a single block or in spaced pieces over weeks. The block group’s scores declined in a straight line. The spaced group’s stayed flat, and were still ahead two years later with nothing further taught. Kerfoot later founded a spaced-learning company, so weight his trials however you like. The shape of the two lines is the point. One is a course. The other is a schedule.

What the wrong unit costs

Saks and Belcourt asked training professionals at 150 organisations how much of their own training was still being applied. Sixty-two per cent immediately. Forty-four at six months. Thirty-four after a year. These are the people who built the programmes, estimating their own results, and two thirds of it is gone inside twelve months.

That is not a content-quality problem that happened to strike 150 different organisations at once. It is what the container does to whatever you put in it.

Why the industry kept the wrong unit

Nobody at an LMS vendor disputes any of this. The course survived for a simpler reason: it was the only unit anyone could afford to produce.

Building around intervals instead of events means every concept needs four or five real question variants rather than one, released over weeks. Questions become the primary content instead of the thing at the end. A single subject produces several times the item volume of a traditional module. At instructional-design rates that math has never closed, so the industry standardised on what it could ship: one module, one quiz, one green tile.

That constraint is what Brik is built around. A source document — an SOP, a chemical safety sheet, a brand standard — goes through our pipeline and comes out as objectives, short cards and item banks with enough variants to run real revisit schedules, delivered on the phone the frontline worker already carries. We still call them courses on the invoice, because that is the word buyers have. Inside, the unit is the objective and the dates it comes back.

One question to ask any vendor

Not what’s in the catalogue. Ask this: what happens on day 50?

If the answer is that the learner can retake the course, the product is event-shaped and the 34% is already built in. If the answer is a schedule, you are looking at something built for memory rather than for the audit.

Sources

• Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17(3), 249–255.

• Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354–380.

• Kerfoot, B. P. et al. (2007). Journal of Urology, 177(4), 1481–1487; Kerfoot, B. P. (2009). Journal of Urology, 181(6), 2671–2673.

• Saks, A. M., & Belcourt, M. (2006). An investigation of training activities and transfer of training in organizations. Human Resource Management, 45(4), 629–648.