The Speaking Training Gap: Why Most L&D Speaking Programs Don't Work

Speaking is one of the few skills that almost every L&D leader will agree matters — for managers, for sales teams, for leadership pipelines, for anyone customer-facing. Yet the way most organisations actually train it hasn't moved much in twenty years: a one-day workshop, maybe two for senior leaders, and then nothing until the next budget cycle.

The data on this is unusually clear. Speaking is a performance skill. Performance skills don't develop from a single event. And every dollar spent treating speaking training as an event is a dollar competing with the part of the brain that forgets 70% of new information within 24 hours.

This post pulls together what the research actually says about how speaking skills get trained inside organisations, why the dominant model under-performs, and what a practice-based alternative looks like for L&D teams ready to move on from the workshop-and-hope approach.

1. How Common Is Speaking Training, Really?

The recognition is universal. The delivery is patchy. A ZipDo analysis found that 52% of organisations provide some form of public speaking training, typically positioned as leadership development. At the same time:

• 75% of employees say their communication skills are insufficient for career advancement.
• 62% of executives cite presentation skills as the most important competency for leadership.

The market is also growing. The speech and presentation coaching market was valued at USD 5.67 billion in 2024 and is projected to reach USD 9.77 billion by 2033 — a compound annual growth rate of 6.2%. Communication-focused training accounts for roughly 2.1% of the estimated $380 billion global L&D spend.

So demand is real, the spend is non-trivial, and the strategic case for speaking training is settled inside most organisations. The harder question is whether the dominant delivery model produces results that justify the investment.

2. The Dominant Model: Once a Year, Then Forget

Most organisations treat speaking skills as a training event, not a practice. A one-day workshop. Maybe two days for senior leaders. Then nothing until next year's budget cycle.

The research on that approach is uncomfortable:

• 82% of people believe they could improve their public speaking skills, but only 11% actively seek training. (ZipDo, 2025)
• 80% of speakers prepare their content thoroughly, but only 50% practise delivery. Content gets the attention; the actual skill of speaking does not. (ZipDo, 2025)
• Employees forget up to 70% of new information within 24 hours and 90% within a month if there is no reinforcement. This is the Ebbinghaus Forgetting Curve, validated repeatedly since 1885 and confirmed in modern workplace studies. (Murre & Dros, 2015)
• One-off workshops raise awareness but rarely build competence. Soft skills training frequently fails to produce the behaviour changes managers expect at work — researchers call it the "soft skills transfer problem." (European Journal of Work and Organizational Psychology, 2024)
• Blended programs (combining in-person, virtual, and self-paced learning) show 34% greater retention of speaking techniques at six months compared to virtual-only, and 12% greater retention compared to in-person-only. (Journal of Applied Communication Research, 2024)

The pattern is clear. Organisations invest in a training moment. The moment passes. The skill fades. And next year they do it again, often with the same provider and roughly the same content.

3. The Practice Problem

Speaking is a performance skill. Like playing an instrument or hitting a golf ball, it improves through repeated, deliberate practice with feedback. It does not improve by attending a lecture about it.

The data supports this directly:

• Practice can reduce public speaking anxiety by up to 68%.
• Practice sessions that simulate actual speaking conditions increase confidence by 30%.
• Yet only 50% of people who prepare a speech actually practise delivering it.

The gap between knowing what good speaking looks like and being able to do it is a practice gap, not a knowledge gap. Workshops can close the knowledge gap in a day. They can't close the practice gap, because that closes through reps, not slides.

Employees pick up roughly 70% of their skills through on-the-job experience rather than formal classroom training. For a skill like speaking — where muscle memory, comfort with silence, vocal control, and audience awareness all matter — that ratio is probably even higher. A two-day workshop cannot replicate what regular, low-stakes practice builds over weeks and months.

4. What L&D Teams Are Spending (and Wasting)

U.S. corporate training expenditure totalled approximately $102.8 billion in the 2024–2025 period, according to Training Magazine's 44th annual Industry Report. Companies spent an average of $874 per learner, and employees received an average of 40 hours of training per year.

The uncomfortable part:

• 75% of training managers are unsatisfied with their organisation's L&D strategy.
• Online training courses see 30% lower completion rates due to lack of engagement.
• 49% of workers admit to clicking through mandatory training just to mark it complete.
• Only 56% of organisations say they can measure the business impact of learning.
• Fewer than 5% of L&D initiatives make it to the measurement stage at all. (LinkedIn 2024 Workplace Learning Report)

The investment is real. The measurement, mostly, is not. And for speaking skills specifically — where the training model is dominated by occasional workshops with no follow-up practice infrastructure — the waste is likely even higher than the average.

5. The Shift That Needs to Happen

The evidence points in one direction: speaking skills development needs to move from event-based training to continuous practice with measurement.

The contrast looks like this:

	Workshop Model	Practice Infrastructure
Frequency	1–2 times per year	Ongoing, weekly or fortnightly
Format	Classroom or virtual session	Short, focused practice reps
Feedback	Trainer feedback on the day	Continuous, data-driven feedback
Retention	70% forgotten within 24 hours	Reinforced through repetition
Measurement	Satisfaction surveys	Skill progression over time
Cost profile	High per-session cost	Lower per-practice-rep cost
Scalability	Limited by facilitator availability	Scales across teams and locations

The returns are there when the training model works. A 2024 McKinsey talent effectiveness review found that organisations with formal communication skills training programs achieved sales conversion rates 18% higher than comparable firms without such programs. An ATD survey from the same year found that companies investing in comprehensive communication skills training reported 29% higher employee productivity.

Both numbers are big. Neither one is achievable with a single annual workshop, because both require the underlying skill to actually transfer to the job — and skill transfer takes practice.

6. What "Practice Infrastructure" Actually Looks Like

Most L&D leaders nod at "practice over events" and then ask the harder question: what does that look like in practice? Translating the research into something operationally workable usually means five things:

1. Short, repeatable reps. Practice attempts of 60–180 seconds, against realistic prompts (status updates, executive briefings, client pitches, all-hands openings). Short reps fit into a working day. Long ones don't.
2. Spaced repetition. Three or four reps a fortnight beats one big effort once a year — precisely because that's the cadence the forgetting curve responds to. The Journal of Applied Communication Research's 34% retention gain at six months was driven by this.
3. Instant, criteria-based feedback. Pace, filler words, vocal energy, eye contact, structure. Surfaced in the same session, so people iterate while the attempt is still fresh, rather than waiting a week for a coach's comments.
4. Skill measurement that isn't a satisfaction survey. Score trajectories, attempt counts, pace and filler-word density over time. The point isn't to police people; it's to give L&D leaders something to put in front of a CFO that isn't "94% of attendees would recommend the workshop."
5. Calendar-friendly. If the practice infrastructure depends on facilitator availability, it isn't infrastructure — it's another workshop in disguise. Asynchronous-first is the only realistic way to scale across teams and time zones.

This is essentially the same logic that's already won in adjacent areas of L&D: sales enablement moved from one-off training to continuous practice with platforms, language learning moved from classroom to spaced-rep apps, and even compliance training is shifting toward microlearning. Speaking has been one of the slowest skill areas to follow that pattern, mostly because of how hard high-quality feedback used to be to deliver at scale.

That constraint has now changed. AI feedback that scores delivery on consistent criteria across thousands of attempts is the missing piece that turns "practise more" from a slogan into something that fits inside a real L&D budget.

7. Where Elqo Sits

Elqo is built for the gap between "we run a workshop once a year" and "we have ongoing, measurable practice infrastructure."

What that looks like for an L&D team:

• Scenario libraries that map to real work. Status updates, exec briefings, sales calls, all-hands openings, performance conversations, customer pitches. Cohorts practise the actual conversations they have, not generic public speaking drills.
• Short, repeatable practice attempts. 60–180 seconds per rep, so a manager can fit three reps into the gap between two meetings without rescheduling anything.
• Instant AI feedback on verbal and visual delivery. Pace, filler words, vocal clarity, eye contact, facial expression, body language — analysed together in a single session.
• Cohort-level analytics. Practice frequency, attempt counts, score progressions, filler-word and pace trends over time. The kind of data L&D teams can actually take to a leadership review.
• No facilitator bottleneck. Scales across teams, locations, and time zones without booking calendars or flying in trainers.

The framing is deliberate: Elqo isn't a replacement for great in-person workshops — those still have a role, especially for kicking off cohorts and creating shared language. It's the missing practice layer underneath them, so the skills introduced in the workshop actually compound through the year instead of fading inside a month.

The Bottom Line

The research is consistent: practice beats passive learning, spaced repetition beats one-off events, and measurement beats satisfaction surveys. Elqo gives L&D teams a way to deliver all three for speaking skills, without the scheduling constraints, facilitator bottlenecks, and measurement blind spots that plague the workshop model.

The question for L&D decision-makers isn't whether speaking skills matter. The data settled that years ago. The question is whether the current delivery model is producing results that justify the spend. For most organisations, the honest answer is no — and that's exactly the gap practice infrastructure is designed to close.

Sources

• ZipDo, Public Speaking Statistics (2025)
• Business Research Insights, Speech-Presentation Coaching Market Size (2024–2033)
• DataIntelo, Public Speaking Training Market Research Report (2025)
• Training Magazine, 2025 Training Industry Report (44th Annual)
• High5 Test / Training Orchestra, Employee Training Statistics (2024–2025)
• Ebbinghaus Forgetting Curve, validated by Murre & Dros (2015), PLOS ONE
• LinkedIn Workplace Learning Report (2024)
• European Journal of Work and Organizational Psychology, soft skills transfer review (2024)
• Journal of Applied Communication Research, blended program retention study (2024)
• McKinsey & Company, Talent Effectiveness Review (2024)
• Association for Talent Development (ATD), Communication Skills Training Survey (2024)
• Learnlight, Communication Skills Training for Employees (2025)