A free 60-second check

Do I even need to build a course?

A lot of "training needs" turn out not to be training problems at all. Answer seven quick questions and find out whether a course is the right next step, or whether a checklist, a clearer set of expectations, or a small fix to the workflow would do the job better.

Why ask before you build

AI can draft a course in minutes. Aiming at the right thing is the harder part, and the one a tool can't decide for you.

Modern AI tools, Slate included, have made building a course much faster than it used to be. What no tool can do is tell you whether building a course is the right move in the first place. That decision shapes everything else: scope, format, whether the course actually moves anything on the job. The question worth asking first: is a course actually the thing that will help?

The honest answer is often no. People who study workplace performance keep finding the same pattern: most of the time, when someone is not doing something well, the cause is not a skill gap. The instructions are unclear. The tool is in the way. They do this once a year and forget. The setup quietly rewards the shortcut. None of those problems get solved by a course, no matter how good the course is.

When you ship a course to fix a non-course problem, three things tend to happen. The problem stays. People assume the learners just did not pay attention. And next time you suggest training, no one trusts that it will help, even when it would.

The 60-second check below is based on two well-known ideas that have been around for decades: Robert Mager and Peter Pipe's "is it really a skill problem?" flow, and Thomas Gilbert's checklist of the things around a person that affect how well they work. You do not need a learning-design background to use it. Answer honestly about a specific situation, get a verdict, and decide what to build, or not build, with more confidence.

The 60-second check

Seven quick yes-or-no questions. Your verdict appears once they're all answered.

Your 60-second check

Pick a specific situation: one course you are thinking about, one team, one task. Answer each question honestly. Your verdict shows up once all seven are done.

1There is something specific your learners cannot do well right now.
2Even with their best effort today, they could not get this right on their own.
3They have done this correctly before, but stopped.
4They already know what "good" looks like, in plain terms.
5They have the tools, time, and access they need.
6Nothing about their setup quietly makes the right way harder than the wrong way.
7A short checklist, video, or one-pager would do the same job.

Answered 0 of 7

What to do when a course is not the answer

Six fixes that often beat a course when the real problem is in the setup rather than in the people. Loosely grouped by the area each one fixes.

Information
Spell out what "good" looks like
Write down, in one page, what doing this well actually looks like. Share it with the people who would take the course. Walk through one real example. A surprising number of "training needs" disappear once the standard is finally clear.
Information
Build a job aid or micro-resource
A checklist, decision tree, one-pager, short video, or scannable mini-resource people can reach right when they need it usually beats a course for tasks done weekly or less. A QR code on the wall pointing to a 30-second reference often outperforms a 30-minute course no one remembers. Job aids replace the need to memorize, instead of trying to build memory.
Resources
Fix the tool or the workflow
If the tool is broken, slow, or missing, training cannot make up for it. Look at the workflow first. Sometimes the right answer is a better template, a fixed form, or a single button, not a course.
Incentives
Check the rewards and consequences
If doing it the right way is slower, harder, or quietly penalized, a course will not change behaviour. Look for goals that pull in opposite directions, unclear ownership, or unspoken trade-offs that make the wrong way the easy way.
Knowledge
Show, do not train
For things people do rarely, a 15-minute walkthrough from someone who already does it well usually outperforms a course. Save the course for skills people use often enough to actually remember.
Capacity
Look at the role itself
If the same gap shows up across lots of different people in the same role, the role may be the real problem. Tweak the job description, the way you hire, or how the work is split before you commit to a course.

Six signs you might be about to build the wrong thing

If two or more of these are true, slow down and look at the problem one more time before you start the course.

1. You were handed the answer, not the problem

Someone said "make a course on X" without describing what they actually want to change. Ask what would look different if the course worked. If the answer is fuzzy, the course will be too.

2. There is no way to tell if it worked

If you cannot name one thing that would happen differently afterward (fewer mistakes, faster onboarding, more sign-ups, fewer support tickets), the course isn't aimed at a real target. Pin one down before you start.

3. You already trained on this, and nothing changed

If a previous course covered this and the behaviour did not change, another course is unlikely to help. Something else is in the way. Look for that first.

4. People do this once or twice a year

For things people do rarely, even a great course fades by the time they need it. A short reference they can pull up in the moment will outperform it. Build that instead.

5. The real fix is a message, not a lesson

Sometimes the only gap is that nobody told the team about a new policy, tool, or expectation. A short note, a quick all-hands, or a manager 1:1 beats a course for one-time announcements.

6. The goal is just "raise awareness"

Awareness on its own is not really a course. If you only need people to know that X exists, write a clear announcement and a one-pager. Save the course for cases where you actually need them to do something differently.

Where these questions come from

Two well-known ideas, in plain English, that most do-I-need-training questions still trace back to.

Mager and Pipe: is it actually a skill problem?

Mager and Pipe, 1970

Robert Mager and Peter Pipe wrote a short book in 1970 with a single very useful question at its centre: is the person not doing this because they cannot, or because something else is in the way? If they cannot, training might help. If something else is in the way, training will not.

Their best-known test: "Could they do it if their life depended on it?" If the honest answer is yes, the issue is not skill. It is the setup around them. Expectations are off, the tool is broken, the reward structure is upside-down, or they only do it once a year. A course will not move any of that.

If you like flowcharts, Cathy Moore's Will training help? flowchart is the modern, visual version of the same idea.

Gilbert: most "skill" issues are not about skill

Gilbert, 1978

Thomas Gilbert spent decades studying why people do or do not do their jobs well. He came back with a simple idea: most of the difference comes from the things around a person, not from the person. The right information at the right time, the right tools, and a setup that does not punish doing it well. Together those matter more than how much someone has been trained.

The practical takeaway: if you fix the instructions, the tools, or the rewards around a job, you usually get a bigger improvement than any course can deliver. Training is one lever. It is not the biggest one.

The "what to do when a course is not the answer" list above is loosely grouped by the areas Gilbert pointed to. When the 60-second check leans away from a course, those are the first places to look.

Common questions

Quick answers to the things people usually ask before they decide to build a course.

It is a structured way of asking: do my learners actually need a course, and if so, a course on what? A good needs analysis looks at the gap between where people are and where you want them to be, what is causing that gap, and whether a course is even the kind of thing that could close it.

Yes, especially. If you are coming from Canva, PowerPoint, or a blank page and trying to figure out whether a course is the right next step, this check is built for that decision. You do not need an instructional design background to use it.

A job aid is anything people can look at right when they need it: a checklist, a decision tree, a one-pager, a short video, a labelled diagram. It replaces the need to memorize the steps. Job aids are often the right call for things people do rarely, or where the cost of getting it wrong is high enough that nobody should be relying on memory.

A course pays off when there really is a skill gap, the task happens often enough that people need to internalize it, expectations are already clear, the tools and incentives line up, and a one-page reference would not be enough. Onboarding, complex software, regulated work, and judgment-heavy decisions usually fit.

They are based on two well-known frameworks in the field. Robert Mager and Peter Pipe wrote the foundational book on the "is training the right fix?" question in 1970. Thomas Gilbert later mapped out the environmental factors (information, tools, incentives) that affect performance more than skill alone does. The questions on this page distill both into something you can answer in a minute.

If training is the answer, or a micro-resource is

Build it in Slate, course or job aid

Slate is an AI-powered eLearning authoring tool. When the check above points to a full course, these features help you ship something tight, measurable, and easy to update. When it points to a job aid, the same tool ships a 30-second scannable reference just as well.

Get to a draft fast

Slate's AI turns a prompt or outline into a tightly scoped first draft in seconds. It is a starting point you refine in the editor, not a finished course. Pair it with the check on this page to keep the scope honest.

Built-in knowledge checks

Every lesson can include short questions with AI-generated answers, so practice and retrieval are baked in by default rather than something you have to remember to add.

Know if it actually worked

Export to any LMS as SCORM, xAPI, or cmi5 and get question-level results back. That way you can tell whether the course moved the thing you cared about, instead of just whether people clicked through.

Lots of "needs a course" is really "needs five minutes"

Most training requests turn out to be a five-minute lesson, not a thirty-minute course. For the why and how, see our microlearning guide.

Sanity-check the length first

Use our free seat time calculator to estimate how long the course will actually take to complete. A course that should be five minutes does not need to be forty-five.

Slate as a job aid (with a QR code)

Sometimes the right answer is a single Slate lesson posted as a QR code where the work happens. A focused video, decision tree, or checklist creators can scan and read in 30 seconds, hosted on a public preview link or a password-protected Share & Track link, and updated centrally any time the policy or tool changes.

Try Slate free Read the microlearning guide

Sources and further reading

Mager, R. F. and Pipe, P. (1997). Analyzing Performance Problems: Or You Really Oughta Wanna, 3rd edition. A short, plain-language book on figuring out whether a problem is really a skill problem.
Gilbert, T. F. (1978, reissued 2007). Human Competence: Engineering Worthy Performance. A detailed look at the things around a person, information, tools, rewards, that affect how well they do their work.
Cathy Moore, Will training help? flowchart . A visual decision tree for working out when training is the right answer and when it is not.
Association for Talent Development (ATD) . A professional community for people who design and deliver workplace learning.
The microlearning guide. A short, research-backed guide to designing focused lessons that are five to ten minutes long.
The seat time calculator. A free tool that estimates how long a course will take learners to complete.