Writing Learning Objectives You Can Actually Measure
"Understand" and "be aware of" can't be measured, so they can't guide design. Here is a simple formula, a verb list based on revised Bloom's taxonomy, and before-and-after examples for writing objectives that drive assessment.
A measurable learning objective describes an observable action the learner will perform, the conditions they'll perform it under, and the standard that counts as success. If you can't picture someone watching the learner do it and saying "yes, that's correct," the objective isn't measurable yet.
That matters because objectives are the hinge of the whole course. They decide what the assessment checks, which practice activities you build, and which content you can cut.
The formula: action, condition, standard
This structure comes from Robert Mager's work on performance objectives and still holds up:
- Action: what the learner does, expressed with an observable verb.
- Condition: the situation, tools or information they have.
- Standard: how well, how fast or how accurately.
Example: Given a sample customer email (condition), the support agent will identify the three signs of a phishing attempt (action) with no more than one miss (standard).
Not every objective needs all three parts spelled out for learners. But you should know all three when you design the assessment.
Verbs that work, by level
Revised Bloom's taxonomy (Anderson & Krathwohl, 2001) arranges cognitive processes from simpler to more complex. Use it to pick a verb that matches what the job actually requires:
- Remember: list, recall, name, identify.
- Understand: explain, summarize, classify, compare.
- Apply: use, demonstrate, calculate, complete, execute.
- Analyze: differentiate, diagnose, troubleshoot, prioritize.
- Evaluate: judge, critique, justify, recommend.
- Create: design, draft, build, propose.
Notice "understand" appears as a level but not as a usable verb. "Explain" and "classify" are the observable ways someone shows understanding.
Before and after
Before: Understand the company's data retention policy.
After: Given a customer request to delete their data, choose the correct retention action from the policy and justify the choice.
Before: Be aware of AI-powered phishing threats.
After: Differentiate a legitimate executive request from an AI-generated impersonation attempt using the three-point verification check.
Before: Learn how to use the new CRM.
After: Create a new client record, including all required fields, in under five minutes without help.
Each "after" version tells you immediately what the assessment looks like. That is the test of a good objective.
Match the level to the job, not to ambition
A common mistake is reaching for higher-level verbs to make objectives sound impressive. If the job requires someone to follow a checklist correctly, "apply" is right, and "evaluate" would mean you're testing something the job doesn't demand.
The opposite mistake is more common in compliance training: testing recall ("Which year was the policy introduced?") when the job requires judgment ("Is this situation a reportable incident?"). Recall questions are easy to write, but they rarely predict performance.
Write the assessment item immediately
My rule: every objective gets a draft assessment item within the same working session. If I can't write a realistic question or task for it, the objective is too vague or doesn't reflect real work. This also keeps the course honest; objectives that never get assessed tend to grow content that never gets used.
For behavior-focused objectives, scenario questions usually work better than definition questions. My post on writing better quiz questions with AI covers how to draft those faster.
Where to start
Pull up the objectives from your most recent course. Highlight every instance of "understand," "know," "learn" or "be aware of." Rewrite each one using the formula above, then check whether your current assessment actually tests the new version. In most audits I've done, at least a few objectives turn out to be untested.
Sources
- Lorin W. Anderson and David R. Krathwohl (eds.), A Taxonomy for Learning, Teaching, and Assessing: A Revision of Bloom's Taxonomy of Educational Objectives (Longman, 2001)
- Robert F. Mager, Preparing Instructional Objectives, 3rd ed. (CEP Press, 1997)