You paid for the training. The feedback forms were warm. Three weeks later the new process is still not happening, and the spreadsheet everyone was trained on looks exactly like it did before.
The pattern has a name: the forgetting curve. Commonly cited research finds more than half of new material fades within 24 hours without reinforcement (Ebbinghaus forgetting curve, first published 1885), and it turns $102.8 billion of US training spend (Training magazine, 2023 Training Industry Report) into the most expensive leak in learning retention.
This guide explains what the curve shows, why good training fades even when learners loved it, the five reinforcement levers that beat it, a 90-day plan that costs minutes per learner, and how to prove learning retention stuck — with numbers you can defend.
Key Highlights
- $102.8 billion: US training spend in 2023, about $874 per learner (Training magazine, 2023).
- Commonly cited research puts more than half of new material gone within 24 hours without reinforcement (Ebbinghaus forgetting curve, first published 1885).
- The commonly cited pattern leaves roughly a quarter retained at one week and about a fifth at one month — one-and-done training is a subscription to forgetting.
- Managers drive 70% of the variance in team engagement (Gallup, 2015); their prompts are the cheapest reinforcement you own.
- 94% of employees say they would stay longer at a company that invests in their learning (LinkedIn Learning, 2018).
- In our planning model, spaced refreshers hold about 80% at day 30 versus roughly 20% with no follow-up (illustrative).
- Reinforcement costs minutes per learner; a second training day costs the whole team again (author's position).
What the Forgetting Curve Actually Shows
The forgetting curve is the pattern Hermann Ebbinghaus documented in the 1880s: memory for newly learned material decays sharply in the first hours and days, then flattens. Modern replications keep landing on the same shape, even though the exact percentages vary by material and method. The honest way to use it is as a shape, not a stopwatch.
Commonly cited versions of the curve put retention near 50% within 24 hours, around 25% within a week, and roughly 20% after a month without reinforcement. Different studies land on different numbers, so treat these as direction, not gospel. Direction is all a training owner needs: the cost of doing nothing is most of your program.
| When | Commonly cited share retained | What it means for your program |
|---|---|---|
| Immediately after | 100% | The end-of-course high is real but temporary |
| 24 hours | ~50% | Same-day and next-day reinforcement matter most |
| 1 week | ~25% | Week-1 refreshers decide the trajectory |
| 1 month | ~20% | What survives is what got used |
Key takeaway: treat the curve as a shape, not a stopwatch, and budget reinforcement against the first month.
Why Training Fades Even When Learners Loved It
Enjoyment is not retention. A workshop can score 4.8 out of 5 on the smile sheet and still leave nothing behind, because the fade is driven by four mechanical failures, not by boredom.
- No retrieval. Learners never recall the material after the course; the brain treats it as unneeded and prunes it.
- No spacing. Everything arrives in one sitting, so memory holds one fragile trace instead of several reinforced ones.
- No manager follow-up. Nobody asks about the training again, so nobody retrieves it, and the daily workload wins.
- No context match. The examples came from someone else's industry, so nothing on the job triggers recall.
The fourth one is the quiet killer. If your compliance examples come from a generic vendor library, your team's actual work never cues the memory, and the cue is what saves learning from the curve. Fix the examples before you add more hours.
Key takeaway: the fade is mechanical, which means it is fixable with mechanics, not motivation.
The Five Reinforcement Levers That Beat the Curve
Five levers reliably slow the fade, and none of them require another training day. Spaced retrieval quizzes force recall at day 1, day 7 and week 4, job aids put the answer where the work happens, and manager prompts turn one-on-ones into retrieval events.
Application assignments make the learner use the skill on a real task within a week, and peer teaching is the cheapest lever of all. Stack them and the shape of the curve changes.
In our planning model for a typical skills program: no follow-up leaves about 20% at day 30, one mid-course refresher lifts it to around 55%, and spaced refreshers plus manager prompts hold near 80% (illustrative midpoints; your numbers will vary with topic and team).
| Lever | What it looks like | Owner | Minutes per learner |
|---|---|---|---|
| Spaced retrieval quiz | 5 questions at day 1, 7 and 30 | L&D | 6 per touch |
| Job aid | One-page card in the workflow | L&D + manager | 4 |
| Manager prompt | Two questions in the next one-on-one | Manager | 15 |
| Application assignment | One real task using the skill in week 1 | Manager | 30–45 |
| Peer teaching | Ten-minute huddle, learner explains | Team | 20 |
Key takeaway: stack at least three levers; one refresher alone buys you a week, not a quarter.
Designing Learning Retention Into the Program
Design the reinforcement before the course, not after the smile sheets. The pattern we use with US mid-market teams is a 90-day arc: same-day recap, day-1 quiz, day-7 scenario, week-4 spot check, and manager prompts running the whole time. A quiz touch runs about 6 minutes per learner, a job-aid review 4, a manager prompt 15, and a peer huddle 20 (illustrative planning midpoints).
Total the touches and the learner spends under an hour on reinforcement across 90 days. Compare that to re-running a workshop day for the same team and the arithmetic makes the decision for you. For topics that suit short formats, our microlearning guide shows how to split the course itself into units that repeat cleanly.
Key takeaway: plan the 90-day arc on the same page as the course, or the course page is all you will have.
What the Curve Means for Your Training Budget
Put the fade into dollars and the priority order becomes obvious. US organizations spend about $874 per learner per year on direct training (Training magazine, 2023), and the commonly cited curve says most single-event learning is gone inside a month. The cheapest response is not a bigger budget; it is spending a small slice on reinforcement instead of another delivery day.
In our budget model for a 100-learner program: a workshop day runs about $300 per learner, a quarter's reinforcement touches about $25, and manager follow-up time roughly $40 (all illustrative midpoints). Reinforcement lands near a tenth of the delivery cost and does more for what sticks.
If leadership wants a return number rather than a cost picture, connect the spend to behavior and results using the training ROI formula. The honest version: reinforcement does not add cost so much as rescue cost you already approved.
Key takeaway: spend a tenth on reinforcement before you spend anything on a second delivery day.
How to Prove Learning Retention Stuck at 90 Days
Measurement closes the loop, and the 90-day check is where learning retention is proven or lost. Run a baseline skill check before the course, a scored check at the end, and a same-format check at day 90. The day-90 number is the only one leadership should care about.
In our planning model, both routes start at a 45% baseline and finish the course near 85% (the refresher route reads 86%). At the 90-day check, no reinforcement lands near 35% while spaced refreshers hold 75% (illustrative midpoints).
| Check | When | What it proves | Who sees it |
|---|---|---|---|
| Baseline skill check | Before the course | Where the team started | L&D + manager |
| End-of-course check | Last session | What the course delivered | L&D |
| Day-30 spot quiz | Week 4 | Whether spacing is working | L&D + manager |
| Day-90 skill check | 90 days | Whether it stuck | Leadership |
The day-90 result is also your behavior evidence: if managers see the skill being used and the check scores hold, you have retention; if scores sag while completion was 100%, you had attendance. For the full evaluation ladder, our Kirkpatrick model guide maps what to collect at each level.
Key takeaway: run the day-90 check in the same format as the baseline, or the comparison means nothing.
What Your Business Should Do
Days 1–30: Audit what already faded.
- Pick the last two training programs you ran and list what changed afterward.
- Ask five managers what they still see being applied from each program.
- Baseline one skill with a short check the team can repeat at day 90.
Days 31–60: Build the 90-day reinforcement arc.
- Schedule day-1, day-7 and week-4 retrieval touches for your next program.
- Write the two manager-prompt questions for one-on-ones before launch week.
- Place job aids in the workflow where each skill actually gets used.
Days 61–90: Run the checks and report the honest number.
- Run the day-90 check in the same format as your baseline.
- Report retention alongside completion so the gap is visible to leadership.
- Fold what worked into your standard program template for every future launch.
The Bottom Line
The forgetting curve is not a metaphor; it is the reason $102.8 billion of annual US training spend (Training magazine, 2023) under-delivers. Commonly cited research puts more than half of new material gone within 24 hours, and a one-and-done program leaves roughly 20% at day 30. A 90-day reinforcement arc of quizzes, job aids and manager prompts holds about 80% in our planning model at a tenth of the delivery cost (illustrative). If you want this mapped to your own team, book a free strategy call or start with the free resource library.
Work With Zohaib Butt
1. Free Resource Library — Zohaibbutt.com/store
World-class training and coaching resources, all in one place. Download the assessments, templates and playbooks you need today.
2. Online Courses — learning.zohaibbutt.com
Self-paced leadership and coaching programs for managers and owners who want structure, not scattered tips.
3. Free Strategy Session — funnelsthatsell.com
Book your complimentary strategy call and we will map your biggest people-development bottleneck to a 90-day fix.
FAQ
What is the forgetting curve in simple terms?
It is the finding that memory for new material decays fastest right after learning, then levels off. Commonly cited versions of Ebbinghaus's work put retention near 50% within 24 hours and roughly a fifth after a month without reinforcement. For a training owner, the practical meaning is simple: whatever is not retrieved, used or reinforced in the first week is mostly gone.
How much do learners forget within 24 hours of training?
The commonly cited figure is more than half within 24 hours (Ebbinghaus forgetting curve, first published 1885). The exact number varies by material, method and learner, so treat it as direction rather than a precise measurement. Direction is enough for planning: schedule same-day and next-day reinforcement, because that is when the steepest part of the curve does its damage.
How do you improve learning retention after a workshop?
Run a 90-day reinforcement arc: a same-day recap, a day-1 retrieval quiz, a day-7 scenario, a week-4 spot check and manager prompts in one-on-ones throughout. Add a job aid where the work happens. In our planning model this holds about 80% of material at day 30 versus roughly 20% with no follow-up (illustrative). The levers cost minutes per learner, not another training day.
Does microlearning help beat the forgetting curve?
It helps where it makes repetition affordable. A six-minute unit is cheap to re-deliver at day 1, day 7 and week 4, so spaced retrieval actually happens instead of being postponed. Short formats do not change the curve's shape by themselves; the spacing and the retrieval do. Microlearning is the delivery shape that makes the spacing affordable; pair the two and the fade slows.
