Training sticks when something in the working week that follows it requires the skill, and it gets forgotten when nothing does. The room is rarely the problem. The fortnight afterwards is, and it is the part almost nobody plans. PowerProv gets asked how to make training stick by L&D buyers who have already run good sessions and watched them evaporate by the end of the month.
- Training usually fails after the session, not during it. The work environment is the deciding factor.
- The famous 'only 10% of training transfers' statistic has no research behind it. The underlying problem is real; the number is invented.
- Design for practice, not exposure. People retain what they did, not what they watched.
- One focus skill beats ten good ideas. Teams keep one thing, so choose it deliberately.
- Manager reinforcement in the first fortnight does more than anything else on this list.
- Measure a behaviour four to six weeks later. A satisfaction score on the day answers a different question.
Why does most workplace training get forgotten?
Because nothing in the returning week requires the skill, and unused skills decay.
Picture the normal sequence. A team spends a day on communication. Everyone leaves energised. They return to 240 unread emails, a deadline that moved while they were out, and a manager who wasn't in the room and doesn't know what was covered. Within four days the training is a pleasant memory competing with actual work. Nobody made a decision to abandon it. It simply never came up again.
The three failure points, in the order they usually bite:
- Nothing changed in the environment. Same meetings, same pressures, same incentives, no prompt to do anything differently.
- The manager wasn't part of it. Managers set what gets attention. If the skill never appears in a one-on-one, it isn't a priority and everyone can tell.
- Too much was covered. A day with eleven models introduces zero habits. Attention is finite.
This is not a motivation problem, which is what it gets mistaken for. People who forget a workshop are not lazy. They are responding accurately to an environment that never asked them to change.
What does the research say about training transfer?
That transfer depends on three things, and only one of them happens in the training room.
The foundational work is Timothy Baldwin and Kevin Ford's 1988 review in Personnel Psychology, which pulled together the empirical literature into a model still used today. Their framework identifies three inputs to whether learning makes it to the job: trainee characteristics (ability, motivation, attitude), training design (content, method, how much practice), and work environment (support, and whether there is any opportunity to use the skill). A later review, Transfer of Training: 1988–2011, traces how the field developed from there.
The practical reading of that model is uncomfortable for buyers: two-thirds of what determines transfer sits outside the vendor's control. A provider can nail design. It cannot make your Tuesday different.
The statistic to stop repeating
You will see it everywhere: "only 10% of training transfers to the job." It is not a research finding.
The number traces to a 1982 article by David Georgenson, who used it as a rhetorical opener, describing what training directors say they estimate rather than reporting any measurement. Researchers Robert Fitzpatrick and Will Thalheimer have both documented the trail, and Thalheimer's conclusion is blunt: the 10% number should be ignored.
Worth knowing for two reasons. First, if a training provider quotes it at you, they haven't checked their own evidence, which is a useful signal about the rest of their claims. Second, the real problem is genuine even though the figure is fabricated. Transfer is hard. It just isn't 10%, because nobody measured it.
“It's a safe, fast-paced, mind-expanding environment and gets you thinking differently about how to solve problems.”— Kamal S., Management Consultant
What makes training stick?
Five conditions. Miss the last two and the first three won't save it.
| Condition | What it means in practice |
|---|---|
| One focus skill | Name the single behaviour the session is meant to change. Everything else is context |
| Practice, not exposure | Participants do the thing repeatedly, get it wrong, get coached |
| Relevance people recognise | Exercises reflect situations the team genuinely faces |
| Manager reinforcement | The skill appears in team meetings and one-on-ones within two weeks |
| A reason to use it | Something in the normal week now calls for the behaviour |
The first three are the provider's job, and any decent provider delivers them. The last two are the buyer's job and they are where most programs quietly fail. Choosing a better vendor cannot fix a work environment that ignores the skill.
Why does practice beat presentation?
Because information and capability are different things, and only one of them survives a busy fortnight.
A person who has watched a demonstration of how to handle a defensive colleague knows what good looks like. A person who has attempted it three times, fumbled once, and been coached through the recovery has begun building something they can retrieve under pressure. The second person has a skill. The first has a memory of a slide.
This is why interpersonal skills resist classroom delivery specifically. Communication, listening, thinking on your feet and handling a room are performed in real time with another person reacting to you. There is no version of that which can be transferred by explanation. It has to be done.
It's also why the format of the training day matters more than its length. Six hours of presentation with a ten-minute exercise is a long presentation. Three hours where participants are working for most of them is training.
How does the format change the odds?
Different formats solve different transfer problems. Match the format to the problem, not to the budget line.
| Format | Best for | Transfer strength |
|---|---|---|
| Half-day workshop (3–4 hrs) | A single focus skill, a team that needs a shared starting point | Good, if a manager reinforces it |
| Full day (6 hrs) | Deeper work including innovation and problem-solving | Better, more practice repetitions |
| Six-week embedded class (2.5 hrs weekly) | Making a behaviour change hold | Strongest, because repetition is built in |
The embedded class exists precisely for the transfer problem. A team practising weekly for six weeks has to retrieve and apply the skill between sessions, which is the mechanism a one-off day can't reproduce. Each week the work environment gets tested and adjusted rather than left to chance.
None of these substitutes for another. A half-day gets a team moving quickly and gives everyone shared language. Six weeks changes how the team operates. The common mistake is buying the first and expecting the second.
What should a manager do in the first two weeks after?
Three specific things, none of which take long.
- Name the skill in the next team meeting. Out loud, on the agenda. Two minutes. This single act tells the team the day mattered.
- Ask about it specifically in one-on-ones. Not "how was the training", which gets "good". Something like "where did you try the listening thing this week, and what happened?"
- Use it visibly yourself. A manager who never demonstrates the skill has answered the question of whether it matters.
There is a fourth, and it applies to whoever signed the invoice: schedule a check at week six before the training happens. A calendar entry made in advance is the difference between a review and an intention. Leaders building the case for spend in the first place will find PowerProv's piece on leadership buy-in useful alongside this.
What should you measure?
A behaviour, at a distance from the event, against something you knew beforehand.
The default measure is the post-session feedback form, which reliably tells you whether people enjoyed the day. Worth collecting and not the same question as whether anything changed. Three measures that carry more:
- Demonstration at four to six weeks. Can people do it, not describe it.
- Unprompted appearance in normal work. The skill shows up in a meeting nobody framed as a test.
- A before-and-after measure of the skill itself. This is rarer than it should be, because it requires the provider to have measured something before training started.
That last one is what the Personal Power Index™ is. It is PowerProv's ongoing longitudinal study, measuring a set of skills before and after each workshop across hundreds of participants since it launched in 2023. It reports capability change, not satisfaction, which is a deliberately harder thing to report.
PowerProv, Personal Power Index™
The 82% figure is worth a second look, because it is the one that shows the measurement is real. Decision-making and leadership shift more slowly than listening or confidence, and a provider reporting the same number for everything is describing marketing rather than measurement. Buyers evaluating providers on this basis will find the questions worth asking a useful companion piece, along with the ROI question.
PowerProv's own method is built around this problem rather than around content delivery. Participants spend the session doing instead of watching, which is what survives the walk back to the desk. The improv label misleads a lot of buyers, so it's worth being plain: there is no performance, no audience, and nobody has to be funny. It is not a comedy workshop. What people practise is listening properly and responding without a script to a real professional situation, which is the behaviour Monday will ask for.
“Exceptional professional development and everyone walked away inspired, enthused and enlightened.”— Siobhan M., LEAD Project Officer
Bottom line
- The room is rarely the problem. The fortnight afterwards decides whether anything holds.
- Ignore the 10% statistic. It was a rhetorical question in 1982, never a measurement.
- Buy practice, not presentation. Interpersonal skills can't be transferred by explanation.
- Book the manager, not just the workshop. Reinforcement in the first two weeks does more than a better vendor will.
PowerProv runs workshops for teams of 12 or more across Australia and APAC, in person and online, rated 4.9 stars over 13 years with organisations including Google, Westpac, PwC, Bain & Co and Canva. Every workshop carries a money-back guarantee, and investment is quoted as indicative ranges because group size, facilitators and travel all move the number.
Book a discovery call to talk through what needs to change, or see how it works first.
Sources
- Transfer of Training: 1988–2011, with the Practitioner in Mind · ERIC / ProQuest
- Georgenson and the origin of the 10% transfer claim · Work-Learning Research
- Personal Power Index · PowerProv
Frequently asked questions
Why does workplace training get forgotten so quickly?
Because the working week usually gives people no reason to use it. Training transfer research going back to Baldwin and Ford in 1988 points at three factors: who the learner is, how the training was designed, and the work environment they return to. The third is the one organisations control most and manage least. A skill nobody is asked to use decays.
How do you make training stick?
Design for practice rather than exposure, pick one focus skill instead of ten, and change something in the fortnight after so the skill gets used. The most reliable single lever is manager reinforcement: if the skill never appears in a one-on-one or a team meeting, the workshop was an event, not a change.
Is it true that only 10% of training transfers to the job?
No, that figure has no research behind it. It traces to a 1982 article by David Georgenson, who posed it as a rhetorical question about what training directors estimate, not as a finding. Researchers Robert Fitzpatrick and Will Thalheimer have both documented that no data supports it. The underlying concern is real, but the number is not evidence.
Does spacing training out work better than a single session?
For most behavioural skills, yes. Skills built through repeated use across weeks tend to hold better than the same content delivered once, because each return visit forces retrieval and application. That's the logic behind PowerProv's six-week embedded class, where a team practises weekly rather than in one hit.
What should a manager do after a team training session?
Three things in the first fortnight. Name the focus skill out loud in the next team meeting so it stays on the agenda. Ask about it in one-on-ones, specifically rather than generally. And use it visibly yourself, because a manager who ignores the skill has told the team what it's worth more clearly than any email.
How do you measure whether training stuck?
Measure a behaviour at a distance from the event. A satisfaction score on the day measures enjoyment. Useful measures include whether people can demonstrate the skill four to six weeks later, whether it shows up unprompted in normal work, and whether a before-and-after measure of the skill moved at all.
Does experiential training stick better than classroom training?
For interpersonal skills, generally yes, because the learner does the thing rather than hears about it. Nobody learns to handle a difficult conversation from a slide. PowerProv is built on that principle, and its Personal Power Index™ measures the change before and after each workshop rather than asking whether people enjoyed the day.
