Throw Out Your Training Request Form

If your training request form is the first place a performance problem enters L&D, I think you should get rid of it.

I do not mean the form you use to book a known program, arrange mandatory training, or enroll someone in a certification. Those are delivery requests. Keep them.

I mean the form a manager uses when something is wrong and L&D is being asked to help.

A few weeks ago, I wrote about why L&D should diagnose before it prescribes. There is a practical consequence to that argument that deserves more attention.

If the first thing a manager sees is a form called Training Request, the process may already be pointing them toward the answer.

Open your form and look at the first five questions

Do they sound like this?

• What training do you need?
• Who is the audience?
• What are the learning objectives?
• What format do you prefer?
• When do you need it?

Those are useful questions once training has been established as the right response.

Before that, they are specifications for a solution.

The manager has effectively diagnosed the problem and prescribed the treatment before L&D has examined anything.

That is backwards.

The cost is not just the course

Suppose a business unit has a quality problem and asks for a two-hour refresher for 2,000 employees.

That request commits 4,000 employee hours before anyone counts design time, facilitation, vendor fees, administration, manager follow-up, or the work those employees are not doing while they are in training.

Maybe the refresher is exactly what is needed.

But what if the problem is a new workflow that creates errors? A system people are working around? Conflicting instructions from managers? A performance measure that rewards the wrong behaviour?

The organization can spend 4,000 hours and still own the problem.

Then the course closes, completion reaches 96%, and the dashboard turns green.

Activity has been completed. The original performance problem may not have moved at all.

This is why I see training intake as a resource-allocation decision, not an administrative step.

We already know the principle

The CDC’s guidance on needs assessment is explicit: understand the current situation first, identify the source of the performance gap, and determine whether training is needed. It specifically tells practitioners to focus on the current situation rather than proposed solutions and to consider causes such as policy, work environment, leadership, communication, knowledge, skill and behaviour. See the CDC guidance.

ATD’s performance-consulting guidance makes the same basic point. When a manager arrives with a solution already in mind, the job is to move the conversation back to expected performance, current performance and the causes of the gap. See the ATD article.

So this is not a new theory.

The more useful question is why our front door can still make the wrong behaviour easy.

Replace the training request with five questions

I would start with a Performance Problem Intake.

1. What should people be able to do?
   Describe the performance you expect to see.
2. What are they doing now?
   Describe the observable gap.
3. What evidence tells us the gap is real and worth solving?
   Use operating data, quality results, customer feedback, errors, observations, system data, or whatever reliable evidence exists.
4. What do you think is getting in the way?
   Treat the answer as a hypothesis, not a diagnosis.
5. What would improve if the problem were solved?
   Connect the change in performance to something the business cares about.

That is enough to start a useful conversation.

The manager does not need to become a performance consultant. L&D can do the diagnostic work.

Maybe people lack knowledge. Maybe they need practice. Maybe the process is getting in the way. Maybe managers are reinforcing the wrong behaviour. Maybe the evidence is too weak to justify an intervention yet.

Find that out before anyone starts writing learning objectives.

Training should have to earn its place

This is not an argument for less training.

If people genuinely need new knowledge, skill, judgement or practice, give them what successful performance requires.

I use the term minimum viable intervention for this. Minimum does not mean cheapest or shortest. It means no more intervention than the problem requires, and no less than successful performance requires.

Sometimes that will be a substantial learning program.

Sometimes it will not be learning at all.

The diagnosis decides.

Keep one request form

There is still a place for a form when the need is already known.

Call it a Learning Delivery Request.

Use it for mandatory training, an existing program, certification, or a capability gap that has already been diagnosed.

Then keep a different front door for unresolved performance problems.

The distinction is simple:

Known learning requirement: help us deliver it.

Performance problem: help us work out what needs to change.

Do not make both enter through a form that already has the word training at the top.

AI makes this more urgent

We can now turn a weak request into a polished course faster than ever. AI can help write objectives, generate scenarios, build assessments and create supporting material in minutes.

That is useful after the decision is sound.

It is dangerous when the diagnosis is weak.

I wrote about that problem separately in AI Can Build the Course. These Five Questions Tell You Whether It Should.

Faster production does not rescue a bad intervention choice. It just gets us there sooner.

Try this with your own form

Take ten minutes and open the form your organization uses when someone asks L&D for help.

Read the first five substantive questions.

Are they helping you understand the performance problem?

Or are they helping you specify training that someone has already decided they need?

If it is the second one, change the front door.

Ask for the problem first.

Then earn the right to prescribe the solution.

Solve the right problem.