Training Needs Analysis Template: Decide What to Train Before You Build Anything

A training needs analysis is the work you do before designing training: establishing which business problem you are trying to move, which people are involved, what they can already do, and where the shortfall sits. The output is not a course list. It is a ranked set of gaps, each with a cause and a proposed response.

The distinction that saves the most money is cause. A gap caused by missing knowledge responds to training. A gap caused by an unclear process, missing tooling, or a broken incentive does not, and building a course for it wastes the budget while leaving the problem in place.

1. State the business problem in a measurable sentence. "Refund disputes are escalating at 22% and we want that under 10% by Q2" is workable; "the team needs upskilling" is not.
2. List the roles and tasks that touch that problem. Keep the unit at the level of a task someone performs, not a broad theme like "communication".
3. Gather evidence of the current level. Use knowledge assessments, work samples, observation, quality or ticket data, and manager input. Two independent sources per task beats one confident opinion.
4. Size the gap. Record the required level, the current level, how critical the task is, and how many people are affected. Multiply criticality by population to get a weighted score you can sort on.
5. Decide the response per gap. Train, coach, change the process, fix the tool, or accept the risk. Assign an owner and a re-measurement date to each decision.

The downloadable template below uses these columns. Keep them; the analysis loses its teeth when the evidence and decision columns get dropped.

Self-assessment alone is the most common failure mode: people who do not know a topic tend to rate themselves generously, and the confident-but-wrong cases are exactly the ones worth finding. Pair every self-rating with an objective source.

Knowledge questions work well for policy, product and process recall. Work samples and observation work well for physical or judgement-heavy tasks. Operational data - ticket reopen rates, defect rates, audit findings - is usually already sitting in a system and costs nothing to pull.

They overlap and are often run together. A skills gap analysis compares current capability against required capability across a population. A training needs analysis starts from a business problem and ends with a decision about what to do, of which training is only one option.

• Starting from a course catalogue and working backwards to justify it.
• Measuring themes ("leadership") instead of tasks anyone can observe.
• Collecting evidence once and never re-measuring, so nobody learns whether the response worked.
• Treating every gap as equal instead of weighting by criticality and population.
• Leaving rows without an owner, which guarantees nothing happens after the spreadsheet is circulated.

A one-off spreadsheet ages fast. Metronic turns the evidence step into something that repeats: upload the material a role depends on, generate assessment questions from it, and the platform tracks per-person and per-topic mastery over time - so the current-level column refreshes itself instead of needing another survey round.