Static Ad Output Template
Provides a structured Markdown template for producing static ad campaign batches with image prompts and ad copy for Meta Ads Manager.
What this file does
Provides a structured Markdown template for producing static ad campaign batches with image prompts and ad copy for Meta Ads Manager.
When to use it
- Running batch static ad campaigns on Meta platforms
- Needing a repeatable format for image prompts and ad copy
- Coordinating image generation with ad copywriting in one document
- Standardizing output across multiple ads or team members
Assumes this stack
Static Ad Output Template
Use this structure for campaign batch outputs.
Contents
- Part 1: Image Prompts - Generate all images first (line 25)
- Part 2: Ad Copy - Copy for Ads Manager (line 70)
- Naming Conventions - File and image naming (line 174)
- Format Pair: 1:1 + 9:16 - Vertical-first design strategy (line 191)
---
type: output
format: static-ad
date: YYYY-MM-DD
status: draft
platform: meta
---
# Campaign Batch {number} — {Campaign Name}
Generated: {date}
Target: {Offer name and price}
Destination: {CTA URL}
**Workflow:** Generate all images first (Part 1), then copy ad text (Part 2)
---
# PART 1: IMAGE PROMPTS
Generate all images first. Design 9:16 vertical first, then center-crop for 1:1 square.
---
## Ad 1: {Angle Name} — {batch#}.1
**Angle:** {description}
**Avatar:** {target persona}
---
### {batch#}.1_IMG_01 — {Descriptive Name}
**Vertical (1080×1920) — design this first:**
```text
{Full prompt for 9:16 vertical. Place critical content (headline, product, key visual) in center 1:1 zone. Fill top/bottom margins with atmospheric/contextual elements.}
Square (1920×1920) — center-crop from vertical:
{Center-crop the vertical to extract the 1:1 safe zone at 1920×1920.}
{batch#}.1_IMG_02 — {Descriptive Name}
[Repeat structure]
{batch#}.1_IMG_03 — {Descriptive Name}
[Repeat structure]
Ad 2: {Angle Name} — {batch#}.2
[Repeat full structure for each ad]
PART 2: AD COPY FOR ADS MANAGER
Copy and paste into Ads Manager after images are ready.
Ad 1: {Angle Name} — {batch#}_IMG_01
Primary 1 — Deep Ad (~500 words)
Hook: {123-135 chars}
{Full primary text}
{CTA URL}
Primary 2 — UGC/Native (~200 words)
Hook: {hook}
{Primary text}
{CTA URL}
Primary 3 — Direct Response (~300 words)
Hook: {hook}
{Primary text}
{CTA URL}
Primary 4 — Pattern Interrupt (~80 words)
Hook: {hook}
{Primary text}
{CTA URL}
Primary 5 — Testimonial (~300 words)
Hook: {hook}
{Primary text}
{CTA URL}
Headlines — {batch#}_IMG_01
Headline 1 — Proof-led
{Headline}
Headline 2 — Mechanism-led
{Headline}
Headline 3 — Outcome-led
{Headline}
Headline 4 — Curiosity-led
{Headline}
Headline 5 — Benefit-led
{Headline}
[Repeat for Ad 2, Ad 3, etc.]
---
## Naming Conventions
**Folder:** `outputs/YYYY-MM-DD-static-ads-{campaign}/`
- Date: `YYYY-MM-DD`
- Type: `static-ads`
- Campaign name: lowercase with dashes (required)
**Batch file:** `static-ads-batch-{###}.md`
- Example: `static-ads-batch-001.md`
**Full path example:** `outputs/2026-01-15-static-ads-january-launch/static-ads-batch-001.md`
**Review log:** `review-log.md` (same folder)
**Image naming:** `{batch}.{ad#}_IMG_{image#}`
- `001.1_IMG_01` — first image for Ad 1
- `001.1_IMG_02` — second image for Ad 1
- `001.2_IMG_01` — first image for Ad 2
---
## Format Pair: 1:1 + 9:16
Facebook Ads Manager accepts exactly two image uploads per ad: **1:1 (square)** and **9:16 (vertical)**. There is no 4:5 option.
### Design Strategy
Design the **9:16 vertical first** with critical content in the **center 1:1 safe zone**. Then center-crop for the square version. One design → two uploads.
### Prompt Approach
1. **9:16 prompt** (1080×1920) — Full creative with critical content centered
2. **1:1 extraction** — Center-crop the 9:16 to 1920×1920
**Standard 9:16 prompt directive:**
```text
Aspect ratio 9:16, resolution 1080x1920. Place all critical content (headline, product, key visual) in the center 1:1 zone. Fill top and bottom margins with atmospheric/contextual elements. Keep text and important elements centered vertically between 25% and 70% from top to stay within safe zones.
Post-Processing (When Using Nano Banana)
If Nano Banana generated the image:
- Raw output is PNG at 1024x1024
- Resize to 1080×1920 (9:16) and 1920×1920 (1:1 center-crop)
- Convert PNG → JPEG, compress under 300KB
- See
image-generation-workflow.mdfor full pipeline
Image Index
When Nano Banana generates actual images, create an image-index.md in the batch folder:
# Image Index — {Campaign Name}
| File | Angle | Style | Format | Prompt Summary |
|------|-------|-------|--------|----------------|
| 001_01_graphic_square.jpg | Authority | Graphic | 1:1 | Bold framework on dark gradient |
| 001_01_graphic_vertical.jpg | Authority | Graphic | 9:16 | Same, full vertical |
| 001_02_lofi_square.jpg | Social Proof | Lo-fi | 1:1 | Casual phone-style screenshot |
| ... | ... | ... | ... | ... |
Generated: {date}
Model: gemini-3-pro-image-preview
Cost: ~${total}
What's inside
2 parts (image prompts, ad copy), naming conventions, format pair guide, image index template
Change this for your project
- Replace
{Campaign Name}with your campaign name - Replace
{CTA URL}with your destination URL - Replace
gemini-3-pro-image-previewwith your image model - Replace
outputs/folder path with your own directory structure
Where it goes
Save as CLAUDE.md in your repository root. Claude Code reads it automatically at the start of every session.
Worth borrowing
- Design 9:16 vertical first, then center-crop for 1:1 square to reuse one composition across two aspect ratios
- Generate all images before writing ad copy to keep workflow sequential
- Use a review log and image index to track batch progress and costs
Related Documents
Code indexing for AI agents: summarization strategies and evaluation systems
Synthesises 2024-2025 research on code indexing for AI agents, covering summarisation strategies, hybrid retrieval architectures, and evaluation benchmarks.
write-script
Guides an AI through writing a YouTube video script with strict structural, language, and quiz rules for a survival channel.
Claude AI Git Workflow Integration
Recommends using the git-ai-commit CLI tool for AI-generated commit messages instead of manual ones.
Missing Business Agents Research — FLUXION 2026
Identifies 12 missing business operations agents for an indie software company and ranks them by impact and effort with €0 implementation plans.