Essay / The 10X thesis

The villains of training: why it fails, and why it has always failed

If traditional management training actually worked, my company would not need to exist. Here is the part I find genuinely fascinating: it has never really worked.

Josh Gain/Founder
July 2026•7 min read
$350bn+ spendBehaviour changeA century of researchTodaythe gap

Have a think about the logic we have all quietly signed up to. A company buys some training, the managers get better, and the problems disappear. That is the deal, and it is the story behind somewhere north of $350 billion of global spend every single year.

The thing is, it does not happen. Here is the bit I find genuinely fascinating though: it has never really happened.

The short version

  • Global spend on training runs north of $350 billion a year, yet the research on the transfer problem shows most of what is taught never changes how people work.
  • This is not a recent failure. Studies going back the best part of a century land in the same place: awareness was never the bottleneck.
  • The model survives on incentives. Workshops are easy to scope, price and invoice. Solved problems are messy, slow and almost impossible to sell by the day.
  • Four villains keep the old model alive: train and pray, the training pile-on, the parachute expert, and make-believe metrics.
  • They are not four failures but one design flaw seen from four angles: the model was built to fill rooms, not to change what happens on a Tuesday afternoon.
  • What works is the opposite design: an embedded Performance Partner using data to find the real problems, community to solve them, and engineered execution to make action the default.

Not a recent failure

It has failed for the best part of a century.

People have been researching what is called the transfer problem, the gap between what someone learns in a training room and what they actually do differently afterwards, for the best part of a hundred years. The findings land in roughly the same place every time: most of what gets taught never makes it back into how people work. Managers turn up, nod along, rate the day highly, and then go back to their desks and carry on pretty much as they were.

$350bn+

spent globally on training every year. The industry has grown almost every year while the evidence that it changes anything has stayed stubbornly thin. If a medicine failed at that rate it would be pulled off the shelves. Training just gets a rebrand instead: new models, new acronyms, new slide decks, same result.

The industry is set up to sell content and events, not to fix problems.

So why does it carry on? Honestly, I think it comes down to incentives. A workshop is easy to scope, easy to price, easy to deliver and easy to invoice. A solved problem is messy and slow and almost impossible to sell by the day. So the industry sells the thing that is sellable, and everyone quietly agrees not to ask too many questions about what happened afterwards.

Before I go further, I want to say something clearly. If you have bought training that did not stick, none of this is a judgement on you. You were sold a model that was never designed to deliver what it promised. The sharpest HR and L&D people I meet worked this out years ago, and their scepticism is completely justified. What they have been missing is not judgement, it is an alternative worth pointing the budget at.

The four villains

Four villains keep the old model alive. Once you spot them you cannot unsee them.

01

Train and pray

Teach a skill in a room, hope it transfers, pray things improve.

The whole traditional model rests on an assumption that sounds mad when you say it out loud. You teach someone a skill in a room, you hope it somehow transfers back into their day job, and then you more or less pray that things improve.

Here is what nobody in the industry wants to admit. Managers do not struggle because they have never heard of coaching, or feedback, or delegation. Ask any workshop whether they know they should delegate more and every hand goes up. Awareness was never the problem.

They struggle because knowing something does not solve the problem in front of them, and because nobody is there when the workshop ends and a manic Monday arrives. No cadence, no accountability, nobody checking in. Just a slow fade back to business as usual, with a workbook gathering dust in a drawer.

The fix

It is not better content. There is more free, world-class content out there than any manager could get through in a lifetime. The fix is what happens after the room: commitments made in front of peers, a rhythm that keeps coming back round, someone whose actual job is making sure the action happened. Development happens through action, and the learning turns out to be a by-product of the performance rather than the other way round.

Our answer → Execution

02

The training pile-on

More frameworks. More models. More work on an already full plate.

Training almost always arrives as an addition to an already overloaded role rather than a solution to it. A manager who is drowning in delivery gets pulled out for two days, handed a new framework, and sent back to an even fuller inbox than the one they left. The moment the content stops helping with the problems actually on their plate, they disengage. Honestly, fair enough.

There is something underneath that as well. Experienced managers do not want to feel like trainees. They want to be trusted to go and fix real problems in their organisation, and if you treat a capable adult like a student you will lose them within the first hour.

The fix

Stop adding and start working on what is already there. Development should run on the real problems sitting on a manager's plate this week, not on case studies about a fictional company. When the work of the programme is the work of the job, you stop having to chase engagement, because it shows up on its own.

Our answer → Real work

03

The parachute expert

“Here is what worked for me, somewhere else. Copy it.”

You will know this one when you see it. A single external guru parachutes in with stories about what worked for them, at a different company, in a different era, with a completely different set of constraints, and the pitch is essentially copy me. Managers spot the disconnect in about four minutes and switch off. I have watched it from the back of the room, and painfully, earlier in my career, from the front of it too.

While all that is going on, the most valuable expertise available never gets touched: the manager two desks away who solved this exact problem last year, in this business, with these constraints. The intelligence is already in the building. The guru model just talks over the top of it.

The fix

Community. Managers learn fastest from each other, that is what we see time and again. Put managers in front of peers, inside the organisation and beyond it, who have genuinely faced the same challenge for real. The job of a decent development partner is not to import answers, it is to unlock the ones already there.

Our answer → Community

04

Make-believe metrics

Attendance, completions, and a happy sheet about the sandwiches.

Think about what training actually measures. Attendance, completions, and a happy sheet on the way out asking whether people enjoyed the day. It is all activity and none of it is impact. None of it tells a CEO whether performance improved or whether a single problem actually went away.

This explains something that has always struck me as odd: development budgets are usually first in the queue when cuts come, despite everyone agreeing that managers matter enormously. It is not that the work does not matter. It is that the return is invisible. Nobody measured the thing that mattered, so when the CFO asks what the spend achieved, the honest answer is a shrug and an attendance sheet.

The fix

Measure movement rather than learning. Get a proper picture of where managers actually are at the start, track what changes, and tie the whole thing to problems the business already cares about. Do that and development stops being discretionary, because the budget holder walks into the next planning round with evidence instead of hope.

Our answer → Data

Four villains, one answer

They are not four failures. They are one design flaw, seen from four angles.

The traditional model was built to fill rooms, and in fairness it does that brilliantly. What it was never built to do is change what happens on a Tuesday afternoon when a manager is staring down a hard conversation they have been putting off for three weeks.

Train and prayThe training pile-onThe parachute expertMake-believe metricsDatafind what to solveCommunitysolve it togetherExecutionmake action the default

One embedded Performance Partner. All four villains answered.

That is why we built 10X Managers the way we did. Not a trainer who turns up, delivers and leaves, but a Performance Partner embedded in the business. The data surfaces what actually needs solving, the community works out how to solve it by drawing on people who have genuinely been there, and then we engineer the execution, through commitments and cadence and accountability, so that taking action stops being the hope and becomes the default.

Training has always failed because it was never really built to succeed at the thing it claims to do. Once you accept that, the way forward gets a lot simpler.

Stop settling for training that ticks the box.

Let us embed a Performance Partner in your business and work on the real problems on your managers' plates.

See how we do it instead →

Or read the client stories where this played out in practice.