7 min read
Beginner
Draft: this article is under editorial review and may change.In plain words: psychomotor verbs are precise action verbs. They name something you could watch being done, such as prepare, rearrange or adapt. You do not need to know the term to use the idea.
If you have written a CLO (course learning outcome)A statement of what a student will be able to do after a course, starting with a measurable verb such as "construct" or "analyse"., someone has probably crossed out your “understand” and written “use an observable verb” in the margin. “Students will understand cash flow” cannot be assessed. “Students will prepare a statement of cash flows that agrees with the closing cash balance” can.
The same rule works on AI. “Help me with my bank reconciliation” gives the AI no idea what finished looks like, so it guesses. “Rearrange my bank-reconciliation worksheet so every adjustment comes before the closing balance” gives it one clear job, and gives you one clear thing to check.
The idea in 30 seconds
Vague verbs, vague answers
"Help", "look at" and "improve" leave the AI to guess the job.
- Vague verbs, vague answers "Help", "look at" and "improve" leave the AI to guess the job.
- Circle your verb Swap it for one you could watch being done: rank, prepare, adapt.
- Verb + object + standard "Rank the five proposals against our four criteria."
- Climb only as high as needed Follow → adapt → design. Higher hands over more judgement.
- Higher means check more A precise verb makes the answer easier to check, not automatically right.
- The verb is the instruction. Everything else is detail.
Key takeaways
If you remember nothing else, remember these.
- Vague verbs such as help, look at, improve and check produce vague answers, because the AI cannot tell what “done” means.
- Precise action verbs such as construct, rank, adapt and trace describe a result you can inspect.
- The verb sets the depth of the task: follow a template, adapt something that works, or design something new.
- The higher you climb, the more judgement you hand to the AI, and the more carefully you check.
Why “help me with” gets you an average answer
Imagine telling a new clerk “handle the claims”. Do you mean check them, sort them, approve them or file them? A good clerk asks. An AI usually does not. It picks a little of everything and carries on.
Take “improve my lab sheet”. That could mean fix the grammar, reorder the steps, add safety warnings or simplify the language. The AI usually does a little of each, and you spend ten minutes working out what changed.
Lecturers already have a test for this: how would I know it was done? If you cannot watch it happen or check the result, the verb is too vague.
| Vague verb | Sharper verb | What you can check |
|---|---|---|
| help with | construct | a complete item in the structure you gave |
| look at | categorise | every item sits in exactly one group |
| improve | revise … so that … | the condition after “so that” is met |
| check | identify | a list of the specific problems you named |
| think about | compare | a side-by-side table with named criteria |
Borrowing a habit lecturers already have
In Malaysian OBE (outcome-based education)The approach used in Malaysian higher education where courses are designed around what students should be able to do at the end. practice, learning outcomes are usually sorted into three kinds, each with its own ladder of verbs:
Three kinds of learning outcome
Think (cognitive)
From remembering to creating. Verbs: explain, compare, evaluate.
- Think (cognitive) From remembering to creating. Verbs: explain, compare, evaluate.
- Do (psychomotor) From noticing cues to designing new ways of working. Verbs: prepare, adapt, design.
- Feel (affective) Attitudes and values. Verbs: value, advocate. These rarely suit AI.
Sources: the thinking ladder is Bloom's taxonomyA ladder of thinking skills, from remembering to creating. as revised by Anderson and Krathwohl (2001). The doing levels most cited in Malaysian practice come from Simpson (1972). The feeling levels are usually credited to Krathwohl and colleagues (1964).
Of course, an AI has no hands. It cannot tighten a bolt or count petty cash. So we are not asking it to perform Psychomotor (domain)Outcomes about doing things with hands or body. skills. We are borrowing the habit behind those verbs: each one names an action you could observe and a result you could inspect.
When the task is thinking rather than producing, such as judging three proposals, a thinking verb (compare, evaluate) fits better. Use whichever verb describes the result most precisely. Feeling verbs translate poorly, since an AI cannot genuinely care about something, so we leave them aside.
The verb ladder: depth lives in the verb
Think of the doing levels as a ladder. At the bottom, a learner notices cues and gets ready. In the middle, they copy a demonstration, then work alone, then work fluently. At the top, they adapt the skill to new conditions and finally design something new.
The same ladder describes how much you are asking of an AI.
Tap a step to see an example prompt that uses its verb.
Psychomotor domain (Simpson, 1972): The psychomotor verb ladder: P1 to P7
Lower steps: the AI follows your pattern.
Higher steps: you hand over more judgement, so you check more carefully.
Three rungs do most of the work in everyday prompts:
- Follow: “Follow our month-end checklist template to write…”. The AI copies your pattern. The result is predictable and easy to check.
- Adapt: “Adapt last semester’s lab sheet to the new PLC training set…”. The AI changes something that already works. You compare it with the original.
- Design: “Design a new petty-cash control for our branch…”. The AI invents. This can be the most useful and is always the riskiest, because there is nothing to compare it with.
The practical rule: climb only as high as the task needs. If you have an approved template, ask the AI to follow it. If you have something that nearly works, ask it to adapt. Ask it to design only when you truly want something new and have time to test it.
A verb alone does not fix the level. Pair every verb with an object and a standard. “Construct three CLOs, each with a verb, a condition and a standard” is far clearer than “construct CLOs”.
Sharpen a prompt yourself
Pick a scenario below. The highlighted words are vague verbs. Tap one, choose a sharper verb and watch the prompt rewrite itself. Each option shows how deep the task becomes.
Notice two things as you play. First, the sharper options are longer: a precise verb pulls an object and a standard in with it. Second, mixing levels in one prompt is normal. “Identify the problems, then revise only those steps” is a simple spotting step followed by an adapting step, on purpose.
Examples from the classroom
Drafting course learning outcomes
AI is good at first drafts of CLOs that follow your format. It does not know your programme outcomes, your accreditation documents or your faculty’s verb list, so give it those and keep the final decision yourself.
You are an OBE curriculum adviser for a Malaysian polytechnic Diploma in Accountancy.
Course: Financial Accounting 2, Year 1. Our CLOs follow the pattern verb + content + condition + standard, and use verbs at the Apply level from our programme list. Sample CLO: "Prepare a statement of cash flows for a small trading company, using the format in the course notes, that agrees with the closing cash balance."
Construct three CLOs for the bank reconciliation topic that follow the sample's pattern. Present a table with the columns CLO, verb, level and how it would be assessed. Do not invent programme outcome numbers or accreditation requirements I have not given you.Read the draft as you would read a colleague’s. Does each verb match the level you intended, and could you actually assess it in class?
Adapting a workshop practical
Workshops change equipment more often than they change outcomes. That makes adapt the natural verb for lab-sheet work.
You are a senior instructor at a TVET college who writes lab sheets for first-year certificate students.
Attached: last semester's lab sheet for our old PLC training set. This semester we use a new training set with a different input/output layout, pasted below.
Adapt the lab sheet to the new training set. Keep the objectives, safety section and marking scheme unchanged. Change only the wiring table and the procedure, start every step with an action verb, and end with a list of every change so I can check each one on the rig.Writing rubric descriptors
Rubric descriptors have the same problem as prompts. “Shows good understanding of cash flow” is hard to mark consistently. Ask the AI to rewrite every RubricA marking guide that lists the criteria and describes what each level of performance looks like. descriptor so it begins with an observable verb and states a standard. For example: “Prepares a bank reconciliation that agrees with the cash book balance, with every adjustment labelled.” Or, in a workshop: “Calibrates the pressure sensor to within the tolerance in the lab manual, without prompting.”
You will still adjust the standards, but the marking discussion with your co-assessors gets much shorter.
Beyond the classroom: SOPs and reports
Industry has used this habit for decades. A good SOP (standard operating procedure)A step-by-step instruction sheet for a routine job. starts every step with a verb an operator can carry out: isolate, inspect, tighten, record. Prompts for SOP work should do the same.
Improve this SOP for the new packing line and check it for safety.
What you typically get back
Smoother wording, a few added lines about 'ensuring safety', and no record of what changed. You cannot tell whether a real safety review happened.
You are a production engineer who writes SOPs for new operators in a Malaysian food-packaging plant.
Below are the SOP from our old line and the start-up section of the new sealer's manual.
Adapt the old SOP to the new sealer in no more than 12 numbered steps, one action per step, each starting with a verb. Identify every step where a hand could enter a moving part and mark it with ⚠. List every change you made in a table: old step, new step, reason.What you typically get back
A short SOP operators can follow, visible safety-critical steps, and a change table your supervisor can approve line by line. Someone still walks the line to confirm it.
Office reports work the same way. “Write about last month’s expense claims” invites a vague essay. Instead: “Categorise every claim by cost centre, rank the five largest cost centres by total, and flag any claim without a receipt number.” That gives you three results, each of which you can check against the spreadsheet.
The verb leads the Mission
If you have read the AIM method guide, you know the Mission says exactly what you want back. The verb leads it. A useful pattern is:
verb + object + standard + format
- Rank (verb) the five research proposals (object) against the four rubric criteria, scoring each from 1 to 4 (standard), in a table with a one-line reason per score (format).
Write the verb first and the rest tends to follow. Once you have typed “Rank”, you ask yourself “rank by what?”, and the standard appears.
Common traps
- Stacking verbs. “Analyse, evaluate and improve my proposal” asks for three jobs in one breath. Split them into steps, and check each before moving on.
- A verb without a standard. “Rank these” lets the AI choose the criteria. Say what to rank by.
- Climbing too high. Asking the AI to “design” a new SOP when you only needed it to “adapt” the old one creates far more to check.
- Trusting the verb instead of the answer. A precise verb makes the answer easier to check, not automatically correct. Check facts, figures and references, follow your institution’s AI policy, and remove personal data before you paste student or client material.
Check your understanding
Nice work!
Lesson complete
You made it to the end. Take the quiz above if you have not yet, for bonus XP.
- 0 XP
0 day streak