Or: Why a learning format sometimes needs a crash test rather than a kickoff.
There are two types of meetings.
Some end with a plan.
Others end with the phrase: “Let’s just get started.”
In Learning & Development, we are surprisingly good at the second category. We are creative, fast, solution-oriented. We can conjure up a training concept in a short time from half a briefing, two stakeholder opinions, and three calendar gaps. Sometimes that even works. And sometimes… well. Sometimes the learning project rolls out as if someone painted the car doors nicely but forgot the brakes.
And then something happens that is almost folkloric in L&D: We discuss content.
“Maybe it was too theoretical.”
“Maybe it was too long.”
“Maybe it needs more interaction.”
“Maybe it’s the target audience.”
Maybe. Maybe not.
The automotive industry would get nervous at “maybe.” Not because there is no creativity there, but because there is a quiet underlying assumption that we in L&D should borrow more often: When something goes wrong, it is rarely bad luck and almost never just a single mistake. It is a system.
Output First, Then Everything Else
In L&D, a project often begins with: “We need training on…”
In the automotive industry, it begins more like: “What should the vehicle be able to do in the end and under what conditions?”
That sounds trivial, but it is the difference between “we are building something” and “we are building the right thing.”
For learning formats, this means: Outcome before content.
Not “participants know…,” but “participants can…” in real everyday situations, under real conditions, with real disruptive factors.
Once you take that seriously, many decisions suddenly change on their own: duration, format, practice component, transfer mechanics. And you realize that the real question is not “How do we make it nice?” but: “How do we make it effective?”
A Learning Project Needs a Crash Test (Not Applause First)
In L&D, we like to celebrate the launch. In the automotive industry, a launch is not a celebration, more a result. Before that come prototypes, tests, stress tests, checks. And above all: an understanding that early failure is cheap and late failure is expensive.
Apply that to learning:
A pilot is not a “small rollout,” but a test lab.
Not: “Do you like it?”
But: “Where does it break? Where do participants stumble? Where do we lose them? Which exercise does not work? Where does transfer fail in everyday practice?”
The automotive industry would never say: “The test driver thought it was nice.”
They would ask: “Did it brake at 120 km/h in the rain?”
We should be just as unromantic with learning formats: Does it withstand everyday reality?
Away from Blame, Toward Causes
When a learning project goes wrong, we in L&D have two reflexes:
Either we look for blame (“the target audience was difficult”), or we quickly do something new (“we build version 2.0”).
Both can be helpful. But both skip the core: Why did it happen?
Not based on feelings, but on evidence.
This is where the 8D logic from industry is a small challenge and precisely why it is so good. 8D is basically the anti-reflex: first describe the problem clearly, then immediate measures, then causes, then actions, then prevention. Step by step, with data, not with gut feeling.
And suddenly “they were not interested” might become:
- Communication came too late,
- Managers did not release time,
- the LMS did not work well on mobile,
- the practical tasks did not fit the job,
- the output was never clear, so transfer was random.
That is no less uncomfortable. But it is much more useful.
Standards Are Not the Enemy of Creativity
In L&D, there is an unwritten law: Too much standardization kills innovation.
In the automotive industry, the principle is more: Standards create reliability, and reliability creates space for innovation.
Checklists, templates, stage gates, definition of done… that sounds like bureaucracy. Sometimes it is. But in the good version, standards are not shackles but guardrails. They ensure that you do not start from scratch every time. And they prevent a learning project from becoming a creative expedition where in the end no one knows how to get back home.
An example: If you send every learning format through three mandatory quality questions…
- Is the output formulated as behavior?
- Is there real application/practice plus transfer?
- Can we demonstrate success and with what?
…then that is not a straitjacket. It is quality management. And yes: That is attractive. Just differently attractive.
The Manager Is Often the Missing Component in L&D
In the automotive industry, you can build the best vehicle, but if the road is missing, it still will not drive.
In L&D, the transfer environment is that road. And the manager is often the section where everything stands or falls: time release, prioritization, expectation, feedback, reinforcement. But we often treat managers like spectators: “Here is the training—have fun.”
The industry would laugh at that. There it is clear: If you want performance, you must build the system around the person.
That means: Managers are not just stakeholders, but co-owners of transfer.
Not with five additional meetings, but with clear, small actions: briefing, expectation, two check-ins, a feedback moment, a visible signal.
Measurement Is Not Control—Measurement Is Navigation
Many L&D measurements are like a speedometer that only shows “good feeling.” Nice, but not controllable.
The automotive industry does not measure because it loves numbers, but because it wants to recognize early if something is tipping. That is the difference between “We will see at the end” and “We see along the way if we are going off course.”
For learning formats, this means:
- Leading Indicators (early): start rate, drop-off points, exercise submission, participation in peer sessions, support tickets
- Learning Indicators: skill demos, rubrics, task quality
- Transfer Indicators: concrete application on the job, manager observation, follow-up after 4–6 weeks
- Impact Indicators (later): KPI proximity, error rate, productivity, time-to-competence
The most important thing is not the perfect metric. The most important thing is a measurement logic that prevents you from realizing only after the rollout: “Oh. That does not work at all.”
A Culture Where You Can Pull the “Andon Cord”
In production, there is the principle: If something is not right, someone can stop the line. Not because you love drama, but because you love quality.
In L&D, however, “stopping” is often taboo. Too many calendars, too many expectations, too much political momentum. So you keep rolling, even though everyone sees that the wheel is wobbling.
What if we allowed ourselves a little industry discipline:
“If a critical quality criterion is not met, we stop—briefly—and stabilize.” Not out of perfectionism, but out of responsibility.
This is not a romanticization of industry. Not everything is better there either. But one thing is very clear there: Quality is not an “extra.” Quality is a system.
And if we are honest in L&D, we are not building engines, but we are building something that must sit just as securely: Behavior under pressure. Decisions in everyday situations. Communication, leadership, compliance, safety, customer contact.
That deserves a bit more crash testing, a bit less “it will be fine.”
And maybe—just maybe—a turtle of all things will help us with that.
