The Forgetting Curve: What Actually Makes Training Stick?
You ran the annual compliance session in March. Attendance was excellent, the deck was thorough, and every box on the completion report turned green.
By April, a teller is doing the thing the training told her not to do. You wonder whether anyone was listening.
They were. Attention was never the problem. What you are looking at is a well-documented property of human memory, and it has a name.
What Is the Forgetting Curve?
In 1885, a German psychologist named Hermann Ebbinghaus ran a long series of experiments on himself and published the result. Memory, he found, decays on a predictable shape: a steep drop almost immediately after learning, then a slower decline that eventually flattens out. Most of what gets lost is lost early.
Researchers replicated his work in 2015 and found the same curve.
“In the replication, retention fell to roughly a third within an hour of learning, and to about a fifth after six days.”
One caveat matters, and we would rather give it to you than have you find it later. Ebbinghaus tested himself on nonsense syllables, strings like WID and ZOF with no meaning attached. Meaningful material holds up better than that. So treat the precise percentages you see quoted around the internet with suspicion, including the popular claim that staff forget 90% of training within a week. What transfers reliably is the shape: fast loss early, and a long tail of whatever survives.
That shape is enough to explain your April problem.
Why Passive Training Falls Off the Cliff Fastest
The single biggest accelerant is passive consumption. A long video, a slide deck, a policy PDF with a quiz at the end. Your staff receive the information and do nothing with it.
Your brain is ruthless about this. Information that arrives once and is never used again gets filed as unnecessary and discarded, which is a reasonable system for getting through a day and a terrible one for annual training. A three-hour session produces a mountain of material the brain has no reason to keep, so it keeps almost none of it.
This is why compliance training has the reputation it has, and why a completion rate tells you so little. We have written before about what one-and-done training actually costs a financial institution. The forgetting curve is the mechanism underneath that cost.
It also explains a frustration familiar to anyone who has watched a universal banker rollout go sideways. Staff were trained. Staff cannot do it. Both things are true at once.
How to Beat the Forgetting Curve
The curve is not a wall. It responds predictably to how training is structured, and four changes do most of the work.
1. Break it into microlearning
Replace the three-hour session with five to ten minute pieces delivered often. Smaller chunks give the brain less to discard at once, and frequency does more for retention than duration ever will. A few short sessions a week beat one long session a year by a distance.
The usual objection is production: nobody has time to build forty small modules instead of one big deck. That changed when turning existing policies and SOPs into training stopped being a manual job, which is also how small L&D teams get out of the content bottleneck.
2. Space the repetition deliberately
Daily drilling is the wrong answer. You want gaps.
When staff are asked to recall something they are on the verge of forgetting, the act of dragging it back strengthens the memory far more than reviewing it while it is still fresh. The difficulty is the point. Space the touchpoints so each one lands near the edge of recall, and the curve flattens with every pass.
3. Make them retrieve, not review
Rereading feels productive and does very little. Being asked a question you might get wrong does a great deal.
Every retrieval attempt is a rep. This is the practical difference between a module someone watches and a game someone plays, and it is why game-based learning outperforms video for anything you need staff to actually do later.
4. Put the knowledge to use
Your brain keeps what it uses. Interactive simulations let staff perform the task rather than watch it, in an environment close enough to the real one that the memory has somewhere to attach. Practicing a difficult member conversation out loud does more for recall than reading a script about it ten times.
And for the things nobody should have to memorize, the answer is to stop requiring memory at all. Answers available in the flow of work mean a staff member with a member in front of them can find what they need in seconds.
From Completion to Competency
Here is the shift worth making. Completion measures whether training was delivered. Competency measures whether it stuck.
An institution working against the forgetting curve will keep producing perfect completion reports and unprepared staff, and the gap between those two numbers is invisible until a member is standing at the counter. Reporting that separates the two is how you find out before they do. It matters most with new hires, where onboarding quality and turnover feed each other in both directions.
First State Bank of Rosemount saw the difference when they moved off the annual model, in what it did to their compliance completion.
The Final Squeeze
Your staff are not forgetting their training because they stopped caring. They are forgetting it because that is what brains do with information they receive once and never use, and a 140-year-old body of research says so.
Work with the curve instead of against it. Short pieces, spaced out, built around retrieval and practice rather than playback, with answers on hand for everything else.
LemonadeLXP was built that way on purpose: microlearning your staff finish in minutes, game-based practice they return to, and a knowledge base that answers the question at the moment it comes up. It is training your team will actually enjoy completing.
