Training Needs Analysis

The three levels of a training needs analysis

The classic framework, going back to McGehee and Thayer's work in the early 1960s, looks at three levels.

• Organizational analysis. Where is the business going and where is performance falling short? New products, new regulations, a new site, rising error rates, poor customer satisfaction. This sets priorities.
• Task or operational analysis. What does a specific role need to do, and to what standard? This produces the list of tasks, skills and knowledge for each job. A current job description and a skills matrix are good starting points.
• Individual analysis. Which people or groups have gaps against that standard, and how large are they?

Many teams jump straight to the individual level with a survey asking employees what training they want. That tells you about preferences, not needs. Start at the top.

How to conduct a training needs analysis, step by step

1. Define the business problem. Write it as a metric: "First-call resolution in the billing queue fell from 74% to 66% over two quarters." Not "agents need billing training."
2. Describe the desired performance. What does a good performer do that others do not? Observe top performers and ask supervisors.
3. Collect data on current performance. Quality scores, error logs, productivity reports, customer complaints, audit findings, assessment results.
4. Find the causes. For each gap, ask whether it is a skill or knowledge problem, or something else: unclear process, poor tools, missing information, conflicting incentives, workload. Only the first is a training need.
5. Prioritize. Rank gaps by business impact and by how many people they affect.
6. Recommend solutions. Training where training fits, and process, tool or management fixes where it does not.
7. Set success measures before anything is built, so you can tell later whether it worked.

The output feeds the analysis phase of the ADDIE model, the most widely used instructional design process. For a ready-made worksheet, use our training needs assessment template.

Data sources and methods

Use more than one source. Each has blind spots.

• Performance data. The most objective source. Quality scores, defect rates, sales conversion, audit results.
• Observation. Watch the work or listen to calls. You will see workarounds nobody mentions in interviews.
• Manager and supervisor interviews. Good for context; biased toward the problems managers find most annoying.
• Employee surveys and focus groups. Good for understanding barriers; weak for judging actual skill.
• Skills assessments. Tests and practical demonstrations show real capability rather than self-ratings.
• Customer feedback. Complaints and survey comments often point straight at the gap.
• Exit and stay interviews. "I never felt trained for this" is a common theme in early leavers.

A skills gap analysis is a closely related exercise, usually broader and more forward-looking, comparing current workforce skills with what strategy will need in two or three years. TNA is typically narrower and tied to present performance.

A worked example

A regional bank's operations center processes loan documents. Error rates on new mortgage files have doubled over six months, to 9%. The operations director asks L&D for a refresher course.

The L&D partner, Tom, runs a short TNA instead:

• Performance data: 70% of errors come from two fields related to a new disclosure form introduced in the spring.
• Observation: experienced processors handle the form correctly. Processors hired in the last six months get it wrong.
• Interviews: new hire training was built before the form existed and was never updated. The procedure document is also out of date.
• Root causes: a knowledge gap for new hires, plus an outdated procedure that experienced staff work around from memory.

Recommendation: update the procedure document, add a one-page job aid, and add a 45-minute module with practice files to new hire training. No refresher for experienced staff. Cost is a fraction of a full course for 140 people, and error rates fall below 4% within two months.

Why TNA matters for high-volume and frontline teams

When an organization trains hundreds of people a year, small mistakes in what gets trained multiply. Ten unnecessary hours in a new hire program for 500 hourly employees is 5,000 paid hours, plus the delay before each person gets to the floor. A missing topic causes errors across every class.

TNA also protects the training budget in general. Learning and development teams that can show a clear line from business problem to training to result get funded. Teams that build whatever is requested lose credibility when the course does not change anything.

Common mistakes

• Treating every problem as a training problem. The most expensive mistake. Ask what else could explain the gap.
• Relying on wish-list surveys. Employees asking for Excel training does not mean Excel is the gap that hurts performance.
• Analysis that takes months. For most operational problems, two to three weeks of focused data gathering is enough.
• No baseline. Without current performance data, you cannot show improvement.
• Doing it once. Roles, systems and regulations change. Repeat TNA for key roles at least yearly and whenever a big change lands.