The Forgetting Curve: Why Employees Forget 90% of Training
In 1885, a German psychologist named Hermann Ebbinghaus sat alone in a room memorizing nonsense syllables. Then he tested himself — after an hour, after a day, after a week. He wanted to know how fast we forget.
The results were sobering. Within 24 hours, more than half of what he'd learned was gone. After a week, roughly 25% remained. What he discovered we now call the forgetting curve — and it remains one of the most important and most ignored findings in corporate training.
#What the forgetting curve actually says
The forgetting curve describes how memory traces decay exponentially without reinforcement. The drop is steepest in the first hours after learning — not after days or weeks.
The rough pattern:
- After 1 hour: ~56% forgotten
- After 24 hours: ~66% forgotten
- After 1 week: ~75% forgotten
- After 1 month: ~79% forgotten
After that, the curve flattens. What's still there after a month tends to stick longer. The problem: most training never gives knowledge enough time to reach that point.
Ebbinghaus worked with meaningless syllables — the hardest possible material to remember. With meaningful, contextualized knowledge, the drop is slower. But the mechanism is universal.
#Why this destroys corporate training ROI
Take a typical compliance training: a 3-hour webinar, a certificate, done. The employee clicks through, passes the final quiz with 82% — and has forgotten half of it two weeks later.
That's not laziness. That's biology.
The training lost to the forgetting curve — not because the content was bad, but because the format ignores how memory actually works. One-time learning without repetition is like pouring water into a leaking container.
This costs real money. Research from the Association for Talent Development estimates that U.S. companies spend over $180 billion annually on training — a significant portion of which produces minimal lasting impact.
#The solution: actively fighting forgetting
Ebbinghaus didn't just describe the problem. He found the answer: spaced repetition — distributed practice.
When knowledge is reviewed at specific intervals before it's forgotten, the curve flattens. Each repetition makes the memory stronger and longer-lasting.
The optimal review schedule looks roughly like this:
- First review: 1 day after initial learning
- Second review: 3 days later
- Third review: 1 week later
- Fourth review: 2–4 weeks later
After four properly spaced repetitions, knowledge tends to consolidate in long-term memory.
Spaced repetition isn't a new trick. Language apps like Duolingo and Anki have used this principle successfully for years. It works equally well for compliance knowledge, product training, and soft-skills development.
#Active recall: repetition isn't enough on its own
How you repeat matters as much as when you repeat.
Passive repetition — re-reading the text, re-watching the video — is significantly less effective than active retrieval: forcing the brain to reproduce the knowledge before seeing it again.
That can be as simple as:
- A short quiz after each learning unit
- Flashcards with a question on one side, answer on the other
- A task that requires applying what was just learned
- A brief "what did I learn?" reflection after completing a module
Learning science calls this the testing effect — the act of retrieving information improves retention more than re-studying the same material.
#What this means for your training design
Concretely:
No training without a repetition plan. Every course should launch with a follow-up schedule built in — automated reminders, short quizzes at 7, 30, and 90 days.
Shorter units, more often. Instead of a 3-hour block: three 45-minute sessions over three weeks. More logistically complex — but measurably more effective.
Verify, don't just track. Completion rates tell you almost nothing. A better metric: how do employees score on knowledge checks 30 days after training? That shows what actually stuck.
Watch out with compliance training: a passing quiz score immediately after a course is not proof that the content was understood — only that it was present for 20 minutes. Check again at 30 days.
#Forgetting is normal — and plannable
The key insight: forgetting isn't failure. It's the default state. The brain actively decides what's important enough to retain. Information that isn't used again gets marked as unnecessary.
Your job as an L&D professional is to send the signal: this information matters. Here it comes again.
That's not a big investment in new technology. It's a decision about how training gets structured.