# Is practice the right fix?

> Practice fixes skill gaps. Check whether the problem is a skill gap or something in the work environment before you build.

- Canonical: https://docs.devlin.ai/design-guidance/is-practice-the-right-fix

A simulation gives people realistic practice on something they need to do. That moves performance only when the problem is a skill gap. Before you build, check that practice is the lever.

## Skill gaps versus everything else
Practice fixes a skill gap: people know what they should do, but cannot yet do it fluently under real pressure. A rep who knows the de-escalation steps but freezes when a caller shouts has a skill gap.

Practice does not fix problems that live in the work environment or in motivation:

- **Missing tools or information.** People cannot find the policy, the system is slow, or the data they need is not in front of them.
- **A broken process.** The handoff is unclear, approvals take days, or two teams own the same step.
- **Incentives that reward the wrong behavior.** Reps are measured on call length, so they rush the customer who needs time.
- **Unclear expectations.** Nobody has said what good looks like, or managers say different things.

When one of these is the cause, more practice leaves the number where it is, and the training gets the blame.

## Spotting it in the request
You rarely need a questionnaire. Listen to how the stakeholder describes the problem:

- "They know what to do, they just don't do it" often points to incentives or expectations.
- "They can't find the answer" points to information or tools.
- "It worked until we changed the system" points to process or tools.
- "New people struggle with it" usually points to a skill gap.

## Raising it with a stakeholder
Say it once, in one sentence, and then help. For example: "Some of this sounds like the refund policy is hard to find, so a quick reference in the ticketing tool would help alongside the practice." Name what would help, then build the simulation if the stakeholder still wants it. Do not refuse to build or hold the work back until the other cause is fixed.

## What to do alongside practice
- Pair the simulation with the fix the environment needs: a job aid, a process change, or a clearer standard.
- Use the simulation to make expectations concrete: its evaluation criteria are a written definition of what good looks like.
- When incentives work against the behavior, tell the stakeholder the practice results may not carry over to the job until the incentives change, and plan to check on-the-job behavior. See [Measuring whether a simulation worked](https://docs.devlin.ai/design-guidance/measuring-simulation-impact).

## Related
- [Writing learning objectives for simulations](https://docs.devlin.ai/design-guidance/writing-learning-objectives)
- [Choosing text or voice for a simulation](https://docs.devlin.ai/design-guidance/choosing-text-or-voice)