Start from what someone should be able to do
Courses built from what the instructor knows are comprehensive. Courses built from an outcome are finishable.
The usual way a course gets made is that somebody who knows a subject sits down and writes out what they know. It comes out in the order they learned it, at the level of detail they find interesting, and it covers everything, because leaving things out feels like doing a bad job.
The result is thorough and largely unfinished by the people who buy it. Comprehensiveness is a property of a reference work. A course is something else: a route from one capability to another, and routes are defined by where they end.
External notes, checklists and prompts reduce the amount that has to stay in working memory; this overview discusses that process as cognitive offloading.
Write the outcome as a task
An outcome is not a topic and not a feeling. It is something a person can do afterwards that they could not do before, stated concretely enough that you could watch them do it.
'Understand analytics' is a topic. 'Set up event tracking on a site and answer a specific question from the data' is an outcome. The difference matters because the second one tells you exactly what belongs in the course and, more usefully, what does not.
For broader evidence-based guidance on teaching and course design, the Carnegie Mellon Eberly Center publishes university teaching resources.
If you cannot describe how you would watch someone demonstrate the outcome, you have written a topic. Rewrite it until there is a verb and an artefact.
One outcome, or a very small number
Courses accumulate outcomes the way projects accumulate scope. Each addition is defensible and the total becomes impossible.
A short course should have one outcome. A longer one might have three or four that stack. Beyond that you do not have a course, you have a syllabus for a term, and it needs the structure of a term — checkpoints, a schedule and a reason to keep going that is not enthusiasm.
Work backwards to the minimum path
Once the outcome is fixed, the design question becomes subtraction rather than addition. What is the shortest sequence of things someone must be able to do to reach it?
Everything that is interesting but not on that path is a candidate for removal. This is where instructors resist, because the removed material is often the material they care most about. It can go somewhere — an appendix, an optional track, a second course — but it should not sit in the main line where it lengthens the route for everyone.
Name the starting point too
An outcome without a stated starting point is half a specification. Courses that do not say who they are for end up serving nobody: too slow for the experienced, too fast for beginners, and both groups conclude the course is bad.
Write the entry condition as plainly as the outcome. 'You should already be able to write a basic query' is more useful than 'some familiarity with databases', because the reader can check it.
Then check the outcome against the time budget
This is the step almost nobody does. Take the outcome, estimate honestly how many hours of practice a person needs to reach it, and compare that to how many hours your audience will actually give you.
For another university perspective on designing learning around outcomes, the Yale Poorvu Center for Teaching and Learning publishes practical teaching resources.
If the outcome needs forty hours and your learners have six, the course as designed will not work regardless of how good the material is. The honest responses are to narrow the outcome or to change the audience. Publishing anyway and blaming completion rates on motivation is the common third option.