The Training Journal L&D Influence Report 2026 has been published, and it is directionally spot on.

The key elements it finds are that L&D needs to be

1. Closer to the business.
2. Evidence that travels.
3. Performance, not content.

None of that is controversial anymore; it is where the field is already heading. The problem, however, is not direction, but translation.

Many organisations are not struggling because they disagree with these ideas. They are struggling because they do not know what those ideas look like when you have to make real decisions about work, ownership, and trade-offs. “Being closer to the business” is not a behaviour but a structural choice requiring change, support, redesign, and a shift in approach.

The questions that move this forward are simple:

• Who owns the outcome?
• Who controls the budget?
• Who decides what stops?

Without those answers, proximity becomes a meeting pattern but not influence. The same applies to evidence. The report correctly points to access, data, time, and permission as the enablers. That is really useful, but those are not enablers in isolation. They also follow a dependency order: access enables data, data enables permission, and permission enables scaling. They are artefacts of how the organisation is governed.

If L&D cannot access decision-makers, it is not a relationship problem. It is positioning. If L&D cannot use data, it is not a capability problem. It is an integration problem. If L&D does not have time to experiment, it is not a prioritisation issue. It is a capacity and demand management problem.

In other words, evidence does not fail because people do not believe in it but fails because the system does not make it usable.

Performance enablement has the same issue. We say “support work in the workflow” and “reduce friction”, and both are grandma and apple pie – obviously correct. But they are not design choices until you define the unit of performance you are trying to change.

A decision.
A task.
A hand-off.
A judgement under pressure.

Until that is explicit, L&D defaults back to programmes, courses, classes, workshops, modules. Not because it wants to but because it has nothing else to anchor to. The shift from direction to design is this.

Not:
L&D should be closer to the business.
But:
L&D is embedded where decisions about performance are made, with shared ownership of outcomes.
Not:
L&D should use evidence.
But:
Impact is defined upfront in business terms, with agreed signals that inform decisions, not reports.
Not:
L&D should enable performance.
But:
Specific workflows are redesigned with the people who own them, so the right action becomes easier than the wrong one.

This is where most transformation efforts stall – they agree on the principles, but they do not change the system. The report is useful because it makes the direction visible, but the next step is harder.

Design the conditions to make those ideas inevitable, not optional or aspirational. That is where L&D either becomes part of how the organisation runs. Or remains a function that supports it from the outside.