Blog

Vague prompt in, vague answer out. Most bad AI answers trace back to a prompt missing one of four things. This is the framework: Role, Task, Format, Constraint (RTFC).

Role — who should the AI act as?

Give it a perspective. “Act as a skeptical editor”, “act as a senior backend engineer”, “act as a first-time customer”. A role narrows the tone and priorities without you having to spell them all out.

Task — what exactly do you want done?

Replace vague verbs with precise ones. Not “help me with this email” — instead “shorten this email to 3 sentences while keeping the ask clear.”

Format — what shape should the output take?

Bullet list, table, JSON, a single paragraph, a script with numbered scenes — say it explicitly. Models default to prose unless told otherwise, and prose is rarely what you actually need for a re-usable output.

Constraint — what are the boundaries?

Length limits, tone, things to avoid, required elements. “Under 100 words”, “no jargon”, “must include a call to action” — constraints are what separate a generic answer from one you can use as-is.

Putting it together

Act as [role]. [Task, stated precisely]. Output as [format]. Constraints: [list].

Example: Act as a skeptical technical reviewer. Review this product description for unsupported claims. Output as a bullet list, one issue per line. Constraints: under 150 words, cite the specific sentence for each issue.

Why this beats “be more specific”

“Be more specific” is true but not actionable. RTFC gives you four concrete slots to fill, and if any one of them is missing, that’s usually exactly where the output falls short.

Midjourney prompts follow a predictable structure. Once you understand the parts, you can build almost any image prompt without guessing.

The basic formula

[subject], [style/medium], [lighting], [composition], [extra detail] --ar [aspect ratio] --v [version]

Example: a lighthouse on a rocky cliff, oil painting, golden hour lighting, wide shot, stormy sea in background --ar 16:9 --v 6

Subject: be specific, not clever

“A dog” gives you a random dog. “A wet golden retriever shaking off water on a beach at sunset” gives you an actual scene. Specificity is the single biggest lever you have.

Style and medium

Name a medium directly: oil painting, watercolor, 35mm photograph, isometric 3D render, pencil sketch. You can also reference an art movement (“art deco”, “ukiyo-e”) instead of an artist’s name.

Lighting matters more than people expect

Try: golden hour, blue hour, harsh studio lighting, soft diffused light, backlit, neon-lit. Lighting changes the mood of an image more than almost any other single word.

Useful parameters

• --ar 16:9 — widescreen, good for banners/thumbnails
• --ar 9:16 — vertical, good for Shorts/Reels covers
• --no [thing] — exclude an element, e.g. --no text
• --chaos [0-100] — higher values produce more varied, unexpected results

A simple starter template

[subject doing something specific], [medium], [lighting], [camera angle], highly detailed --ar 16:9

Fill in the brackets and you’ll get consistently better results than a one-line vague prompt.

Most “ChatGPT prompt” lists are filler. These ten are ones you can copy, paste, and use today — each one solves a specific, recurring task instead of a vague one-off.

1. Turn a messy meeting into action items

Here are my raw meeting notes: [paste notes]. Turn these into a clean list of action items, grouped by owner, with any deadlines called out separately.

2. Rewrite an email to match a tone

Rewrite this email to sound more [direct/friendly/formal], keeping the same information: [paste email]

3. Summarize a long document into 5 bullets

Summarize the following into exactly 5 bullet points, ordered by importance: [paste text]

4. Turn bullet points into a slide outline

Turn these bullet points into a 5-slide outline, one idea per slide, with a suggested slide title for each: [paste bullets]

5. Draft a difficult message

Help me draft a message to [person/role] about [situation]. Be direct but not confrontational, and keep it under 100 words.

6. Compare two options side by side

Compare [option A] and [option B] in a table across these criteria: cost, time, risk, and long-term impact.

7. Turn a question into a checklist

Turn this into a step-by-step checklist I can follow: [paste question or task]

8. Explain something to a specific audience

Explain [topic] as if you were writing for [a new hire / a non-technical manager / a client]. Keep it under 150 words.

9. Find the gaps in a plan

Here's my plan: [paste plan]. What's missing, and what could go wrong that I haven't accounted for?

10. Turn feedback into a revision plan

Here's feedback I received: [paste feedback]. Turn this into a prioritized list of changes to make, starting with the most important.

Why these work

Each prompt gives the model a clear input, a clear output format, and a clear constraint (length, structure, or tone). That combination is what separates a useful prompt from a vague one — not clever wording.