Five Questions Before Anyone Quotes for Training

The number, and what it does not say

The Department for Education's Employer Skills Survey 2024 puts total employer training expenditure at £53.0bn. In the survey's own words, "This equated to a spend of £1,700 per employee, down from £1,960 in 2022 and a 29.5% decrease since 2011." The total came to 111 million training days, or 3.6 days per employee.

Where the money goes inside a budget is a separate piece; this one is about what has to be answered before any of it is committed. What the survey cannot tell you is whether any of those 111 million days were aimed at the right problem. It counts days. It does not count what changed on the Tuesday after.

Five questions before a format is on the table

Our client journey page says the brief is rarely the real problem, and that the free 30-minute call with one of the founders has no deck in it. Here, in one place, are the five questions we think have to be answered before any quote, ours or anyone else's.

They are the specification, not a sales technique. In our manager programme, the behaviours each scenario puts under pressure are agreed with the organisation before the two days are built, and nothing can be agreed if the behaviour has not been named.

• What would someone see a person doing differently, and by when? Not "better leadership" but a behaviour a colleague could count without asking what it means. If the target cannot be stated as something observable, it cannot be measured afterwards, which is where our measurement guide starts.
• What is stopping it now? The COM-B model splits the answer into capability, opportunity and motivation, and training addresses capability. If the managers already know how to hold the conversation and have no hour to hold it in, a course spends the budget on the one condition that was not missing, the trap our piece on managers' time describes.
• What is the baseline? A measurement taken only after the programme is a snapshot, not a change. The same method before and after, or the result cannot be defended.
• Who has to be in the room, and who is sponsoring? The people whose behaviour has to change, not the people who are easiest to release for two days, and a sponsor who will sit in the debrief in person, because adoption holds when they do and not when they delegate.
• How will you know if it did not work? If the honest answer is a satisfaction score, the programme cannot fail, and a programme that cannot fail cannot be shown to have worked either.

Why the last question matters most

The fifth question is the one budgets are most often spent without asking, and we have our own evidence on it. In the randomised 2020 UCL study by co-founder Ben Laumann, 57 participants learned the same feedback model through slides, a video or an online coaching session of up to 20 minutes with a role-play coach, then delivered a feedback conversation that two assessors rated blind. Rehearsal produced significantly better delivery, with effect sizes of d = 0.75 against slides and d = 1.08 against video. That is one short online exercise, not a measurement of our programmes, and it says nothing about retention weeks later.

The part that bears on budgets is the confidence data. Participants rated their own confidence on a scale of 1 to 7 three times: before, after the teaching, and after the simulation. The slides group went from 4.46 to 5.00 to 5.39. The video group went from 4.69 to 5.13 to 5.38. The practice group went from 4.27 to 5.33 to 5.53. The means rose at every step in every group and the rise from start to finish was significant in all three, while the format had no significant effect on that rating (p = .958) and nor did the interaction between time and format (p = .541). Source: Figure 4 of Laumann (2020), UCL, unpublished research paper; the rest of the study is on our research page.

Put those three rows next to the blind ratings and the budget problem is visible: ask people how the training felt and the two formats that produced significantly weaker delivery report a success. An evaluation built on "do you feel more confident?" will approve almost any spend, including spend aimed at the wrong problem. That is why the fifth question has to be answered with an observable behaviour, and why "we will run a survey at the end" sends the conversation back to question one.

A number we had to take down

We have been caught out by a sourceless figure ourselves. Until September this site quoted a different per-employee training spend. Checked at the source, it appeared nowhere except on our own pages, so it came off every page that carried it and was replaced with the official survey figure above.

A number without a source is an assumption in a suit, and briefs are full of them. Every figure in a brief deserves the question we now put to our own.

What happens after the questions

Sometimes the five answers describe a problem a rehearsal-based programme can reach: a conversation managers avoid, a moment where people go quiet, a decision made badly under pressure. Then we would build for it, priced per engagement rather than per head, with the behaviour named, the baseline captured and the re-measure booked before delivery, so that the same task runs before and after, as our piece on why L&D spend rarely becomes behaviour describes. Sometimes the answers describe a workload, a structure or an incentive that no training touches, and we say so, because the promise to say honestly whether we are the right people only means something if the answer can be no.

Either way the £1,700 is better spent after the questions than before them. Our programmes for companies start from the problem, not the format, and a free 30-minute call is the place to start asking them.

Sources: Department for Education, Employer Skills Survey 2024 · Sidestream, The Study Behind the Method · Sidestream, Client Journey · Sidestream, How to Measure Behaviour Change · Sidestream, What Is the COM-B Model? · Sidestream, New Manager Training · Sidestream, Why L&D Spend Rarely Becomes Behaviour