CLEAR, 5W1H and SMART: Checklists for Better Prompts

Study guide

Checklists for Better Prompts: CLEAR, 5W1H and SMART

Most prompt advice gives you a template to fill in. This course gives you something more flexible: checklists you run over prompts you have already written. You will learn four frameworks, CLEAR, 5W1H, SMART and Five S, and how to use them to diagnose why an answer fell short and repair only the part that failed. The prompt stays in your own words. The checklist simply tells you what is missing or vague.

This course is for busy professionals who use ChatGPT, Claude or Gemini at work and want a repeatable way to improve results. You do not need a technical background. Each lesson explains the framework, shows a worked example from a real job, and names a common mistake. By the end you will have a two-moment routine: run 5W1H before you send, and run CLEAR's Reflective check after the answer arrives.

What is a prompt framework?

A prompt is the request you type into an AI tool such as ChatGPT, Claude or Gemini. A prompt framework is a reusable structure for that request: a short checklist of the parts a good prompt contains, usually named with an acronym so it is easy to remember. Why it helps: the AI fills in whatever you leave out with generic guesses. A framework makes sure you include the parts that change the answer, such as who it is for, what you want to achieve and the format you need.

Frameworks make prompts repeatable: you fill the same parts each time, save the prompt as a template and share it with colleagues. They are not magic words: the value comes from the detail you put in each part. They help most when a request is complex or open to interpretation; a quick, simple question can stay short. There are well over sixty named frameworks, and most of them remix the same ingredients: role, task, context, audience, format, constraints and examples.

In this course you learn CLEAR & 5W1H, which is a set of quality frameworks: checklists you run over any prompt to make it better, rather than slots you fill in. It is best for: checking and improving any prompt; popular in education and libraries.

Checklists, not templates

A template gives you blanks to fill for one shape of task. It is fast when the task matches and awkward when it does not, because you end up forcing your request into someone else's structure. A checklist is different. It is a set of questions you run over a prompt you have already written. The prompt stays in your own words, and the list simply tells you what is missing or vague.

The main value of a check is diagnosis. When an answer is weak, you compare your prompt against each check, find the one that failed, and repair that part instead of rewriting everything. A prompt usually contains a task, some context, a format for the reply, and often a length or tone. Knowing these parts helps you see which check applies where.

Running checks is a habit, not a one off event. The more often you walk the list before sending, the faster it becomes, until you are catching gaps as you type. This course teaches four checklists. CLEAR checks the writing, 5W1H checks coverage of the situation, SMART checks the goal, and Five S gives teams a shared version.

CLEAR by Leo S. Lo

CLEAR was published in 2023 by Leo S. Lo, a librarian at the University of New Mexico, and it was written for information literacy rather than for marketing or sales. It has five parts. Concise, Logical and Explicit build the prompt. Adaptive and Reflective judge the answer.

The first three letters all apply before you press send. They shape the prompt itself so the AI has less room to guess. Concise means cut what does not change the answer. Logical means the request follows a clear order. Explicit means say the format, length and tone you want rather than hoping.

Adaptive means you plan to revise. You read the first answer, find the gap, and send a follow up that keeps the same format rather than starting over. Reflective means you judge the answer, not just the prompt. Ask whether it makes sense, whether it could be invented, and whose view is missing before you use it. The two review letters are the ones most people skip. A prompt that is never revised and never checked is only half of the CLEAR method at work. CLEAR works well beside other checklists. Use CLEAR to build and review, and a goal framework such as SMART when the prompt needs a measurable target.

5W1H: who, what, when, where, why, how

5W1H stands for who, what, when, where, why and how. Each question fills a slot in your prompt, and a missing slot is a gap the AI will fill with a guess. The who slot sets the audience, which quietly controls reading level, tone and how much background you need to include. Naming the reader is often the single biggest improvement to a prompt.

The where slot tells the AI the channel, such as email, slide deck or chat message. Channel shapes length and formality more than most people expect. 5W3H adds how much and how many to the six questions. In 2026 studies it was used to structure intent and cut follow-up rounds on ambiguous business tasks.

Use the framework as a check, not a template. You do not need to write the labels in your prompt. You only need to know that every question has an answer somewhere in the text. The fastest habit is to ask which question your prompt did not answer, then add that one line. One added line usually removes one round of follow-up.

SMART goals in prompts

SMART stands for specific, measurable, achievable, relevant and time-bound. It comes from George Doran in 1981 and works as a checklist for the goal slot of any prompt framework. A specific goal names one clear outcome. Improve our newsletter is not specific. Raise the newsletter open rate is specific, because it names the thing that changes.

A measurable goal carries a number. Raise open rate from 25 percent to 30 percent gives the AI a gap to close, so its advice can target that gap instead of offering generic tips. Achievable and relevant keep the goal honest. The target must be realistic with your resources and must matter to the business, or the AI plans work that cannot land or does not count.

Time-bound means a deadline. Within three months sets the horizon, so the AI can sequence actions across weeks instead of listing everything at once. When the goal slot is SMART, the AI gives advice aimed at that target. You can also ask the AI to rewrite a rough goal as SMART, then check the numbers yourself before using them.

Five S for teams

Five S is a simple rulebook for training non technical teams, so everyone writes prompts the same way without a technical background. The first four S are about the prompt itself: set the scene, be specific, simplify language, and structure the output.

The fifth S, share feedback, is the habit most people skip. It means telling the AI what to change and why, instead of accepting a weak answer or starting again. Feedback works best as a revision, not a restart. The AI keeps the context, so the second answer is usually closer to what you need.

When a whole team uses the same rulebook, prompts become easier to review, copy and improve across the department. This shared language means a colleague can pick up your prompt and understand why it is written that way, which shortens handovers and reduces rework.

Running a prompt check in your job

5W1H before sending gives you six quick questions: who, what, when, where, why, and how. Most weak prompts fail at least two of them, so the check catches gaps before the AI fills them with guesses. CLEAR's Reflective check happens after the answer arrives. You ask whether the brief was met, whether anything was invented, whether the tone fits, and whether a colleague would understand it without extra context.

A five-line checklist turns your team's most common prompt mistakes into a reusable guard. It is faster than rereading a long guide and more specific than a generic best-practice list. Checklists pair with templates rather than replacing them. A template like CO-STAR gives structure, while the checklist catches the gaps the template assumes you already know.

Two minutes of checking on an important prompt is usually less work than rewriting a poor answer. The routine works in any tool, including ChatGPT, Claude, and Gemini. The routine scales from one person to a whole team. When everyone runs the same five lines, the quality of prompts becomes consistent across the group.

Frequently asked questions