Prompts
EN

Image prompt Product & Ads Customizable

Four-Panel SNS Ad Grid With Dissonant Copy

A 2x2 grid of four fictional social ads, each in a different color temperature and layout, with one deliberate mismatch per panel. Bold and thin Japanese type contrast against dense and empty space.

Model
GPT Image 2
Type
2 images
Aspect ratio
1:1
Shared
21 Apr 2026
Likes on X
134
Make it yoursSwap in your own details: 3 blanks to fill

Product categoryProduct categoryProduct category

Image prompt

3,925 characters

task: "Generate fictional SNS ad visuals that make people stop scrolling" goal: - "Create a 'moment of dissonance' that stops the scroll" - "Establish professional ad design (contrast and priority of information)" - "Compose so that people want to compare the 4 ads" - "Make a strong visual in the style of 2025 SNS ads seen on TikTok / Instagram" canvas: aspect_ratio: "1:1" layout: "2x2 grid" spacing_rule: "Keep clear borders between the 4 panels while giving the whole a unified feel" reading_flow: "top-left → top-right → bottom-left → bottom-right" attention_design: rules: - "Create one 'point of dissonance' in each ad (copy, composition, whitespace, expression, etc.)" - "Make it read in the order dissonance → understanding" - "Do not fully explain the meaning in the first 0.5 seconds (a design that makes you want to see more)" hierarchy_rules: - "At most 2 lead elements per ad (①main copy ②visual)" - "Always lower the priority of subcopy, CTA, and proof elements" - "Even if it looks information-heavy, keep the gaze flow simple" layout_variation_rules: - "Vary the layout structure across all 4 ads" - "Be sure to include whitespace type, dense type, diagonal composition, and UI type" - "At most 1 with a centered layout" sns_2026_rules: - "A thumbnail composition understandable in an instant, like a TikTok ad" - "An emotion-first visual, like an Instagram Reel" - "Insert small review-style, UI-style, and badge-style elements" - "Make a strong contrast between bold and thin fonts" - "Create contrast through the difference between whitespace and information density" psychology_effects: core: - "Loss aversion / sense of crisis" - "Gap effect" - "Exclusivity" additional: - "Empathy hook" - "Social proof" - "Curiosity gap" - "Self-projection" copy_rules: structure: "Grab attention → gap → benefit → CTA" constraints: - "Short, strong copy in Japanese" - "One strong message per ad" - "Not over-hyped, but stops the scroll" - "Categorical claims of effect are forbidden (consideration of the Pharmaceutical Affairs Law)" - "Excessively fear-mongering expressions are forbidden" tone: - "Has SNS-ad momentum, but not information overload" - "Make the copy look strong through whitespace and contrast" main_copy_rule: - "10–18 characters" - "Readable in one line" - "As a question or assertion" sub_copy_rule: - "Up to 1–2 lines" - "Strictly supplementary" cta_rule: - "Button-style or boxed" - "Placed at the endpoint of the gaze" - "Always make it stand out" buzz_elements: forbidden_words: - "You lose if you don't know this" - "Stop right now" - "99% are doing it wrong" allowed_gap_patterns: - "Looks ordinary but actually…" - "This is not a ○○" - "Looks cute but is strong inside" urgency_options: - "3 days only" - "1 spot left" - "Free → 0 yen" rules: - "Naturally weave one each of gap, exclusivity, and empathy into every ad" - "Prioritize the dissonance → interest flow rather than blatant hype" genres: - "Education (children's cram school)" - "Fitness (dark gym)" - "Fashion (silhouette-correcting outfits)" - "Lifestyle (stress-relief app)" generation_rules: - "The 4 ads completely differ in color, composition, temperature, and amount of information" - "Design the dissonance, but don't make it cluttered" - "Always place the CTA at the endpoint of the gaze" - "Each works on its own, and side by side they make you want to compare" - "A design that achieves information contrast, target resolution, and tone separation" - "Create a flow of 0.5-second dissonance → immediate understanding" output: type: "High-quality SNS ad visual" description: "An SNS ad image with 4 ads combined into one sheet, balancing dissonance and orderliness"

How to use it

The prompt is written for a text-capable image model, so keep the grid structure and the reading order top-left to bottom-right intact. Swap the four genres (children's cram school, dark gym, silhouette-correcting fashion, stress-relief app) for your own product category, and keep one short main line per panel. It was shared as a GPT Image 2 prompt, so check that the Japanese copy renders cleanly before using it.

#Poster#Social Media#Advertising#Text in Image#UI & App Design

Shared publicly on X on 21 Apr 2026. All credit for this prompt and its result goes to its creator. Is this yours? Ask for an update or removal.

3,925 characters

Questions about this prompt

Which AI model was this prompt made with?

It was shared as a GPT Image 2 prompt, and the image on this page is the result the creator got with it. Other models can read the same text, but the result will look different.

Can I change the details in this prompt?

Yes. It has 3 parts you can swap: product category, product category, product category. Open the prompt studio, fill them in and copy your version; anything you leave empty keeps the original wording.

Who made this prompt?

It was shared publicly on X by @uniyume on 21 Apr 2026. The prompt and its result belong to them, and this page links to the original post.

More from @uniyume