Corporate Training Program Template

Start with the problem, not the course. The U.S. Office of Personnel Management (OPM) says a training needs assessment¹ answers two basic questions: what needs to be done, and why isn't it being done now?

The gap isn't always a missing skill. OPM adds that training is sometimes not the best solution and virtually never the only one. A step people rarely perform may need a one-page job aid more than a course.

Then agree with the sponsor on what success looks like. A 2004 U.S. Government Accountability Office (GAO) guide calls for up-front agreement with key stakeholders on what success is and how it will be measured². Write down any result the sponsor was already promised, word for word.

Build the program around the job itself. OPM describes job analysis³ as examining a job's tasks, the competencies they require and how the two connect, and says it can be used to determine training needs. Write each competency as something a person does, such as "receive a delivery against the purchase order." Rate each one on three scales, and don't change the ratings later to fit the modules you have.

• Frequency: how often the task comes up. Rare tasks get little practice at work.
• Criticality: what a mistake costs. It decides where the time goes first.
• Difficulty: how hard the task is to learn. Hard tasks need practice with feedback.

Next, tie each module to the competencies it teaches. A module counts only if someone who attends will be measurably better at that competency afterward.

Look for two gaps: competencies no module teaches, and modules that teach nothing on the list. Keep each untaught competency on the list at zero minutes, then decide whether to add time, handle it outside training or drop it. A welcome or system update may be worth its time, but it isn't coverage.

Count scheduled contact minutes, not days. Split each module's minutes across the competencies it teaches, check the totals against the timetable and sort by criticality. This shows whether time follows the ratings or the material on hand.

In the template's sample program for 34 warehouse team leads, 660 of the 755 scheduled minutes go to a listed competency. No module teaches three of the 14 competencies, including one rated 5 of 5 for criticality. The two least critical competencies get 105 minutes, while the daily forklift check, rated 5 of 5, gets 60. That may be right, since those two are harder to learn, but it should be a decision.

Evaluation plans are often organized around Donald Kirkpatrick's four levels, which date from the 1950s: reaction, learning, behavior and results. Kirkpatrick Partners, which sells certification in the model, advises starting with Level 4, the results the organization wants⁴, and working backward.

OPM's 2011 Training Evaluation Field Guide⁵ is built on the Kirkpatrick levels. Its lists of pitfalls to avoid include considering evaluation only after a program is under way, relying mainly on surveys, and neglecting Levels 3 and 4.

For each competency, name the measure, then ask four questions. A Level 3 or 4 claim holds only if every answer is yes.

1. 1

   Does a reading exist?

   Name the instrument: a scored exercise, an observation checklist or a system report. A satisfaction rating isn't a score.

2. 2

   Does it name the person trained?

   Check what the data is recorded against. A dock turn time belongs to an appointment and a damage rate to a site, so neither shows that one trainee changed.

3. 3

   Does it move only with this skill?

   A damage rate also moves with the season and the carrier mix. Report it, but not as evidence about this competency.

4. 4

   Is there a baseline?

   Without one, an after-training number is just a number. Capture it before the first cohort starts.

GAO's guide warns that isolating the performance improvements that result from a specific training activity is especially difficult². It describes comparing the performance of individuals and work units before and after training. For one person, that needs data with their name on it and a baseline.

Report those two kinds of gap separately, because they cost different things. In the sample plan, 15 promised levels lack evidence today. Of those, 11 wait on a change to what gets recorded, and four need only a baseline due within 24 days.

Write the facilitator guide, participant workbook and slides from the same map, so all three teach one course. The facilitator guide runs module by module: competencies, minutes, a quick check that it landed and what to cut if time runs short.

The workbook holds what learners fill in and take back to work, including any checklist they'll be observed against. Tell them what will be measured; it shows them what to practice. Score exercises against a written standard, as a grading rubric does, and write learning outcomes if your organization asks for them.

Plan what happens around the sessions, too. The Association for Psychological Science, summarizing a 2012 review by Eduardo Salas and colleagues, notes that what occurs before and after training is just as important for program success⁷. The authors recommend a needs analysis and the right learning climate beforehand, and afterward a check that people apply what they learned at work.

Before the first cohort, send the sponsor one page: the levels promised, the levels the data supports and the reason for each gap. After each cohort, correct the timings in the facilitator guide.

The template at the top of this page works through every check on a made-up warehouse program, and River can run them on yours.