Back/Marketing/Claude/Midjourney
BeginnerMarketing

How to Use the Subject-Setting-Style Framework for Professional Midjourney Images

Write more controllable Midjourney prompts by separating the subject, setting, and style, then vary one photographic choice at a time and keep a prompt log that connects the selected image to its intended use and rights.

How to Use the Subject-Setting-Style Framework for Professional Midjourney Images

Ravi compares a bare office-chair prompt with a subject-setting-style prompt, then varies lighting, location, film stock, camera, lens, and aperture to show how photographic vocabulary changes Midjourney output.

Before you start

What you need

  • A concrete use case, audience, and image placement
  • A subject brief with composition and exclusion requirements
  • A setting with location, time, lighting, and mood
  • A style vocabulary drawn from photography or art you may reference
  • Aspect ratio, resolution, brand, disclosure, and licensing requirements

What you’ll make

A selected image set with reproducible prompts, generation settings, comparison notes, and the provenance needed for safe use in the intended product or campaign.

Tools used

Step by step

The workflow

Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.

5 steps

Step01

Adopt the Subject-Setting-Style Framework

Write the subject as a concrete visual specification: object or person, pose, expression, framing, orientation, count, and details that must or must not appear.

Step02

Specify the Style Using Photographic Language

Describe style with visible photographic properties and permitted cultural references. Use camera, lens, aperture, or film language as cues only when you understand the effect you want.

Example prompt
Create a [image type] of [subject and composition]. Setting: [location, time, weather, lighting, background]. Style: [visual properties], photographed with the look of [camera or film cues], [lens and aperture effect]. Target placement: [placement and aspect ratio]. Exclude [artifacts and unwanted details].

If you lack the vocabulary, analyze a licensed reference for visible composition, lighting, color, texture, and depth. Do not ask for a copy of a named artist's style.

Step03

Construct Your Structured Prompt

Combine subject, setting, and style in a fixed order. Add the target aspect ratio, text-safe area, realism level, and exclusions, then save the exact prompt and generation settings.

Example prompt
an empty stylish office chair, overlooking Milan during an autumn raining morning, Fuji Color C200
Step04

Create Professional AI Portraits

For portraits, define age range, expression, gaze, pose, crop, environment, and intended diversity without copying a real person. Inspect skin, hair, eyes, teeth, hands, jewelry, and background details at full size.

Example prompt
Portrait of a fictional [age range] [person description], [expression and pose], framed [crop] in [setting and light]. Style: [photographic qualities], [lens and depth of field]. Natural skin texture and realistic anatomy. Do not resemble a named real person. Exclude distorted hands, duplicate features, false text, and logos.
Step05

Compare and Refine

Generate controlled variations by changing one element at a time, such as setting, light, lens, or film cue. Compare against the placement brief, select deliberately, and record provenance, edits, disclosure, and usage rights.

Example prompt
Act as an expert photographer. I want to recreate an image's style using an AI image generator. Analyze the provided image and describe its key photographic properties.

Break down your analysis into:
1.  **Subject:** Describe the main subject and its composition.
2.  **Setting:** Describe the environment, time of day, and lighting conditions.
3.  **Style:** Suggest a specific camera brand, lens (e.g., 50mm f/1.8), and film stock (e.g., Kodak Portra 400) that would create this aesthetic, and explain your choices.

[paste image or describe it here]

What good looks like

  • The subject and composition support the actual placement rather than merely looking attractive.
  • Prompt variations isolate meaningful choices such as setting, light, lens, or film treatment.
  • People, text, hands, products, and brand details receive task-specific artifact and bias checks.
  • The team records the generator, model version, prompt, references, selection rationale, edits, disclosure, and usage rights.

Build your next product with ChatPRD

Turn an idea into a PRD, user stories, and a plan.

Try ChatPRD free

After the steps

Runbook notes

How to recover when the loop fails and where human judgment helps.

Recover

If it goes sideways

A style reference imitates a living artist or protected brand too closely
Describe the visual properties you need, use licensed references, and obtain legal or brand guidance for commercial use.
Camera or film terms are treated as a guarantee of physical accuracy
Use metadata as a controllable style cue, compare the visible result, and do not claim the image was captured with that equipment.
A polished image still contains distorted anatomy, false text, impossible objects, or biased representation
Inspect at full resolution, test representative subjects, regenerate or edit defects, and reject unsafe or misleading output.
The chosen image crops poorly or conflicts with the product layout
Generate for the target aspect ratio and text-safe area, then test real crops at every intended breakpoint.

Start shipping
better products.

Join 100,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.

Free to start
No credit card
SOC 2 certified
Enterprise ready