The short version
Decades of research on workplace training say skills last when people use them on their own work, spaced out over weeks, with a manager who knows how to coach them. Enjoying a session predicts almost nothing: the correlation between enjoying training and learning from it is .02, according to the research collected in our field study Game Winning Kickoffs.
| # | Finding | Number | Paper |
|---|---|---|---|
| 1 | Enjoying training does not predict learning | .02 correlation (usefulness .26) | Alliger et al. 1997 |
| 2 | Spaced practice beats one big burst | d 0.46 across 63 studies | Donovan & Radosevich 1999; Cepeda et al. 2008 |
| 3 | Being tested beats re-reading | 56% vs 42% remembered a week later | Roediger & Karpicke 2006 |
| 4 | Unused skills fade hard | d −1.4 after more than a year unused | Arthur et al. 1998 |
| 5 | Own scenarios and trained managers hold skills | 117 studies | Taylor, Russ-Eft & Chan 2005 |
| 6 | Managers' coaching skill predicts results | 1,246 pharma reps | Dahling et al. 2016 |
| 7 | The work environment predicts use better than the room | Two studies | Saks & Belcourt 2006; Blume et al. 2010 |
This piece is one finding set from the sales coaching research hub. For the coaching practice built on it, see the sales coaching framework.
Where these findings come from
The kickoff study, run in October 2026 on public sources, needed to know what research says about events shaped like a sales kickoff: a big burst of content, rated on how it felt, followed by a return to normal work. No peer-reviewed study looks at a sales kickoff itself. So the study collected findings about workplace training in general and confirmed every paper exists via its DOI. The seven below are the ones a sales manager can act on. They come from different fields, pharma sales and classroom memory among them, and each is applied here to the sales floor with that gap in mind.
1. Enjoyment does not predict learning
How much people liked a training session correlated .02 with what they learned (Alliger et al. 1997, Personnel Psychology). Ratings of how useful the session was did better, at .26.
Most training and kickoffs are scored on a satisfaction survey. That survey measures something real, and it is close to unrelated to learning. Corporate Visions, quoted in the study, put it well: "Energy is a great outcome for an SKO, but it's a lousy measurement plan."
What a manager does with it: ask reps after any session what they will do differently on which deal this week. That answer is closer to usefulness than a star rating.
2. Spread it out
Across 63 studies, spaced practice beat massed practice, with an effect size of d 0.46 (Donovan & Radosevich 1999; Cepeda et al. 2008). For something to last a year, the best gap between reviews is roughly 5 to 10% of that year, which works out to roughly every three to five weeks.
A two-day event is massed practice by design. The research says the follow-up matters more than the event's length.
What a manager does with it: put the skill on the agenda of the one-to-one in week one, then again every few weeks, on whatever deal is live. The sales kickoff follow-up plan lays out the calendar.
3. Use beats hearing
A week later, people tested on material remembered 56%, and people who re-read it remembered 42% (Roediger & Karpicke 2006, Psychological Science). The re-readers felt more confident.
That last line matters for sales. A rep who has heard the discovery framework three times feels ready, and on this research is likely less ready than one who has been made to use it once and got it partly wrong.
What a manager does with it: in a deal review, ask the rep to say the next question they will ask the buyer, out loud, in their words. Retrieval is the exercise. Telling them the answer is re-reading.
4. Unused skills fade
Skill loss was near zero straight after training and reached d −1.4 after more than a year without use (Arthur et al. 1998, Human Performance).
This is the real research behind the forgetting claims, and it says something more useful than a percentage: the loss depends on use. A skill used weekly is maintained by the using. We call the gap between what a rep learned and what they do on a call three weeks later performance drift (know-do gap vs performance drift).
What a manager does with it: pick the skills that matter most this quarter and make sure each rep uses each one on a live deal every week or two.
5. Own scenarios, trained managers
Across 117 studies, skills held or grew when people worked on their own situations and when their managers were trained on the same thing (Taylor, Russ-Eft & Chan 2005, Journal of Applied Psychology).
Generic scenarios teach a generic skill. A rep's own deal has the real buyer, the real stakeholders and the real risk, and the skill has to fit it. The second half of the finding is the one most programs skip: the manager has to know the method too, or the rep is coached against a different standard on Monday.
What a manager does with it: run the method on the rep's live pipeline, and learn the method yourself before asking reps to.
6. Coaching skill predicts results
Across 1,246 pharma reps, managers' coaching skill predicted goal attainment. Frequent coaching from less skilled managers was linked to worse results (Dahling et al. 2016).
This is the uncomfortable one. More coaching is not automatically better. A manager who coaches often and badly can do harm, and good coaching skill is unevenly spread across any management team. That is the manager capacity problem in one study: there are never enough good managers to coach every rep on every deal.
What a manager does with it: coach on fewer things, better. Use the same structure each time: brief, observe, debrief. The sales coaching framework sets it out.
7. What happens back at work decides it
What happens back at work, before and after the training, is more strongly tied to whether it gets used than what happens in the room (Saks & Belcourt 2006; Blume et al. 2010).
The room gets most of the budget and attention. The work environment, meaning the manager's questions, the pipeline review, what gets praised and what gets asked about, decides whether anything transfers.
What a manager does with it: change the questions in the meetings that already happen. If the new standard is a dated next step with the budget holder, ask for it on every deal in the pipeline review.
A statistic to retire
"You forget 70% in 24 hours, 90% in a week" appears on 23 of the pages our study collected. It traces to Ebbinghaus in 1885: one person, himself, memorizing nonsense syllables. The study rates it misattributed. Other favorites fare no better: "87% of training is forgotten in 30 days" traces to a 1979 article citing an unpublished Xerox study with no timeframe, and "only 10% of training transfers" began as a rhetorical question in a 1982 trade article.
The real findings above make the case without them. Skills fade without use, and spacing, retrieval and skilled coaching on the rep's own work slow the fade.
What it adds up to
Read together, the seven findings describe a coaching job more than a training event. The skill gets used on the rep's own deals, soon and repeatedly, with someone who knows the method and asks good questions. One more result from the study supports it: in a randomized trial, short, repeated conversations between salespeople about technique raised revenue per call by 24%, and the gains outlasted the conversations (Sandvik, Saouma, Seegert & Stanton, Quarterly Journal of Economics, 2020).
- Ask what the rep will do differently, on which deal, this week
- Put the skill in the one-to-one in week one, then every few weeks
- Have the rep say the next question out loud
- Use each key skill on a live deal every week or two
- Run the method on the live pipeline, and learn it yourself
- Coach on fewer things, better: brief, observe, debrief
- Change the questions in the meetings that already happen
That job is where most teams run out of manager time, and it is the job our AI sales experts take on. The rep brings a live deal to an expert such as Caty on a call and talks it through; the expert runs the team's own method on that deal, asks the retrieval questions, and works out the next conversation with the buyer. Typing is the fallback for when a rep cannot talk. Managers get a read across the team from Bryce.
An easy place to start is the next deal review: ask the rep to say the next question they will ask the buyer, out loud, before you offer yours. The kickoff study is the working behind Episode 02 of Show Your Working, live on Wednesday 11 November 2026 with Celeste Berke Knisely of A Sales Growth Company. The same series tested AI call scoring: average error flatters a lazy scorer.
FAQ
What does research say about sales training effectiveness?
It says enjoyment barely predicts learning (correlation .02; Alliger et al. 1997), spacing beats one burst (d 0.46 across 63 studies; Donovan & Radosevich 1999), being tested beats re-reading (56% vs 42% a week later; Roediger & Karpicke 2006), and unused skills fade (d −1.4 after a year; Arthur et al. 1998). What happens back at work predicts use better than the training room does (Saks & Belcourt 2006; Blume et al. 2010).
How quickly do reps forget sales training?
How fast reps forget sales training depends on how much they use the skill. Skill loss was near zero straight after training and reached d −1.4 after more than a year without use, in Arthur et al. (1998). The popular "90% forgotten in a week" figure traces to Ebbinghaus in 1885, one man memorizing nonsense syllables, and does not describe sales skills.
What is the spacing effect in sales training?
The spacing effect is the finding that practice spread over time beats the same practice in one block. Across 63 studies the advantage was d 0.46 (Donovan & Radosevich 1999). For a skill to last a year, the best gap between reviews is roughly 5 to 10% of that year, or every few weeks (Cepeda et al. 2008).
Does more sales coaching improve results?
More sales coaching does not improve results automatically. Across 1,246 pharma reps, managers' coaching skill predicted goal attainment, and frequent coaching from less skilled managers was linked to worse results (Dahling et al. 2016).
How do you make sales training stick?
Use the skill on the rep's own deals within a week, then every few weeks. Ask reps to say what they will do on a live deal, out loud, and resist re-explaining it for them. Make sure managers know the method, and change the questions in the pipeline reviews and one-to-ones that already happen.