BDC training fades when nothing after the class makes agents use it: no practice, no scored calls, a manager who was not in the room, a CRM that pulls the other way, and new hires who never saw it. The research on forgetting is thinner than the famous percentages suggest, but it does point to the fixes.
What the Famous Forgetting Numbers Measured
Every training pitch has the slide: people forget 70% in 24 hours, and 87% of sales training is gone in 30 days. Neither is a measurement of sales training. The first comes from Hermann Ebbinghaus, who in 1885 published experiments on one subject, himself, memorizing lists of nonsense syllables. He measured how much relearning time a list saved after a delay and reported that "after 24 hours about one third was always remembered" (Ebbinghaus, Memory, chapter 7). A 2015 replication by Murre and Dros in PLoS ONE found a similar curve, again with a single subject.
So the curve is about meaningless lists, learned once by rote, by one man. It says nothing about skills.
The 87% figure traces to Neil Rackham. A 2010 slide deck of his has a slide titled "The Impact of Training: Xerox Study" that reads "87% skill loss- within a month!", followed by a line saying managers reinforce new skills through coaching. The deck gives no sample size, year, method or publication, and no published study behind the figure could be located. The slide is a story about one company losing its skill gains when coaching did not follow the course. It is not "content forgotten", not an industry average, and not Ebbinghaus.
A third number, "only 10% of training transfers to the job", has no data at all. Robert Fitzpatrick traced it to a rhetorical question in a 1982 article that cited no evidence and was then repeated as fact (Fitzpatrick, The Industrial-Organizational Psychologist, 2001).
What the Published Research Does Show
There is real evidence on what makes learning last and reach the job. None of it was gathered in a BDC, each finding has a limit, and most were read from the published abstracts, so they are stated no more broadly than that.
| Finding | What was studied | Limit |
|---|---|---|
| Practice spread over time produces better later recall than the same practice crammed together, and the best gap gets longer the longer you need to remember. Cepeda and colleagues, Psychological Bulletin, 2006 | A quantitative review of 839 assessments in 317 experiments | Verbal recall tasks in lab studies, not workplace skills |
| Being tested on material makes it stick better than rereading it, though rereading leaves people more confident. Roediger and Karpicke, Psychological Science, 2006. A later review found tests that make you produce the answer help more than tests where you pick it. Rowland, Psychological Bulletin, 2014 | Two experiments with college students and prose passages, tested 2 days and 1 week later; then a quantitative review | Rereading did better when the final test came only 5 minutes later |
| Behavior modeling training (show the skill, practice it, get feedback) works. Its effect on skills and job behavior held steady or grew over time, while its effect on factual knowledge decayed. Taylor, Russ-Eft and Chan, Journal of Applied Psychology, 2005 | A quantitative review of 117 studies | Effects were largest on learning, smaller on job behavior, smaller still on business results |
Training professionals at 150 organizations estimated that 62% of employees apply training right after a course, 44% after six months and 34% after a year (Saks and Belcourt, Human Resource Management, 2006). Those are trainers' estimates gathered by survey, not measured behavior. The same study found that what happened in the workplace before and after training was more strongly related to transfer than what happened during the class.
The behavior modeling review also found transfer to the job was greatest, in the abstract's words, "when mixed (negative and positive) models were presented, when practice included trainee-generated scenarios, when trainees were instructed to set goals, when trainees' superiors were also trained, and when rewards and sanctions were instituted in trainees' work environments." In a phone room: play a bad call and a good one, let agents bring their own hard callers to role-play, have each agent pick one behavior, train the manager, and make the pay plan reward the trained behavior.
Four Reasons It Fades in a Dealership
Nobody practiced after the class
The class ends on Thursday. On Friday the phones are busy, and under pressure people say what they have always said. A word track tried once in role-play loses to a habit with two years behind it.
The manager was not in the room
The manager was covering the desk or decided the training was for the agents. Now the one person who listens to calls every week cannot tell whether an agent is doing what was taught. The manager coaches the old way on Monday, and the agents conclude that the old way is what counts.
The CRM contradicts the training
The class taught a phone call first, then a text. The CRM's task list still schedules an auto-email and a call on day three. The class taught an appointment with a date, a time and a named person to ask for. The appointment screen still accepts "Saturday" with no time. The pay plan still pays on appointments set, not shown. Agents follow the system that produces their tasks and their paycheck. When the process and the training disagree, the process wins.
The people who were trained have left
Phone rooms turn over. If three of the eight agents on your board were hired after the class, more than a third of the room never saw it, and they are learning the job from whoever sits next to them.
The Fixes, in Order of Cost
- Put practice on the schedule. Ten to fifteen minutes, three or four mornings a week, one scenario at a time. Make agents produce the words from memory instead of reading the script back. Our role-play drills for BDC agents are built for blocks this short.
- Score calls every week. Two recorded calls per agent, marked yes or no against the trained behaviors, shows whether the class is still alive. Use one BDC call scorecard for everyone.
- Coach the call, not the person. A review of 607 effect sizes found feedback improved performance on average, yet more than one third of feedback interventions made it worse, and feedback lost effect as attention moved toward the self and away from the task (Kluger and DeNisi, Psychological Bulletin, 1996). "You did not offer two times on this call" works. "You are not a closer" does not. The weekly rhythm is in the BDC coaching cadence.
- Train the manager first, or at least in the same room. The manager should be able to demonstrate every word track and score a call the way the trainer does.
- Make the CRM agree with the class. Rewrite the task cadence, templates and required appointment fields the same week.
- Give every new hire the same training, on a clock. A written first month with a sign-off before live leads keeps the standard when the faces change.
Expectations, public numbers and consequences are covered in building a BDC accountability culture. The sales floor version of this problem is in why sales training does not stick.
When Fading Is Not the Real Problem
Sometimes the training held and the numbers still fell. Listen to ten calls. If agents are using the trained behaviors and appointments are still short, the leak is somewhere training cannot reach: leads routed to nobody after 6 p.m., two agents covering a workload that needs four, or a sales desk that does not greet the appointments the BDC sets. Train, fix or outsource walks through how to tell a skills problem from a process problem or a coverage problem.
What Pinnacle Does After the Class
We sell training, so weigh this accordingly. The Sales BDC Training page says "We don't believe in train and leave" and lists what follows the initial training: weekly or monthly performance reviews, live call audits and feedback sessions, refresher courses for new hires or underperforming staff, and leadership coaching for BDC managers. The PDS Boot Camp sends each attendee home with a 90-day action plan.
We found no independent study that measures the sales effect of BDC training. Whether it holds in your store is something to measure, not assume. Any trainer's published result ranges, ours included, are claims to test against your own baseline. Measuring BDC training results at 30, 60 and 90 days shows how, and the Sales BDC Training Resource Center covers the steps in between.
Frequently Asked Questions
Is the forgetting curve real?
Yes, as a finding about rote memory. Ebbinghaus measured how fast one person lost lists of nonsense syllables, and a 2015 replication found a similar curve. It was never a measurement of skills training, so the percentages on training slides do not apply to your BDC.
How often should BDC agents practice after training?
Short and frequent beats long and rare. As a practitioner's rule, run a ten to fifteen minute drill three or four times a week for the first month, then at least weekly. Lab research supports spreading sessions out, though it studied verbal recall, not phone skills.
Should the BDC manager sit through the training?
Yes. A manager who cannot demonstrate the word tracks or score a call to the trained standard will coach the old way by default. One review of behavior modeling training found transfer was greater when trainees' superiors were also trained.
How can I tell whether our last training has faded?
Pull five recorded calls per agent from this week and mark the three or four behaviors the class taught: a greeting with a name, discovery questions, an appointment ask with two specific times, a confirmation. If most calls miss most of them, it has faded. Rates alone will not tell you, because lead mix moves them too.
Summary
The famous forgetting percentages do not measure sales training: Ebbinghaus tested one man on nonsense syllables, and the 87% figure has no published study behind it. The published research supports spaced practice, recall, rehearsal with feedback about the task, a trained manager and a workplace that matches the class. In a dealership that means a short drill block, two scored calls per agent each week, a trained manager, a CRM that matches the class, and the same first month for every new hire.
