Introduction
Creating social media graphics manually takes 15-30 minutes per post. For a business posting 3-5 times per week across 3 platforms, that’s 5-10 hours of design work. Automated image generation reduces this to seconds: define a template, feed it data, and generate hundreds of branded graphics.
This post shows you how to automate social image creation with Python.
Approach 1: Pillow (free, local)
Python’s Pillow library can generate images from templates:
from PIL import Image, ImageDraw, ImageFont
import os
def create_social_image(title, subtitle, output_path):
# Create canvas (1080x1080 for Instagram)
img = Image.new("RGB", (1080, 1080), color="#1e293b")
draw = ImageDraw.Draw(img)
# Load fonts
title_font = ImageFont.truetype("fonts/Poppins-Bold.ttf", 64)
subtitle_font = ImageFont.truetype("fonts/Poppins-Regular.ttf", 32)
# Title (centered)
title_bbox = draw.textbbox((0, 0), title, font=title_font)
title_width = title_bbox[2] - title_bbox[0]
title_x = (1080 - title_width) // 2
draw.text((title_x, 400), title, font=title_font, fill="white")
# Subtitle
sub_bbox = draw.textbbox((0, 0), subtitle, font=subtitle_font)
sub_width = sub_bbox[2] - sub_bbox[0]
sub_x = (1080 - sub_width) // 2
draw.text((sub_x, 500), subtitle, font=subtitle_font, fill="#94a3b8")
# Logo/watermark
logo = Image.open("logo.png").resize((100, 100))
img.paste(logo, (50, 50), logo if logo.mode == "RGBA" else None)
img.save(output_path)
print(f"Created: {output_path}")
# Generate 10 images
titles = [
("5 Python Tips", "For beginners"),
("AI Tools Roundup", "March 2026"),
("Freelancing Guide", "Start today"),
]
for i, (title, subtitle) in enumerate(titles):
create_social_image(title, subtitle, f"post_{i+1}.png")
Approach 2: Canva API (designer-quality)
Canva’s API lets you generate images from templates:
import requests, os
CANVA_API_KEY = os.environ["CANVA_API_KEY"]
def create_from_canva_template(template_id, text_replacements):
# Get template
template_url = f"https://api.canva.com/rest/v1/designs/{template_id}"
headers = {"Authorization": f"Bearer {CANVA_API_KEY}"}
# Create design from template
r = requests.post(
"https://api.canva.com/rest/v1/designs",
headers=headers,
json={"design_type": {"type": "preset", "name": "instagram_post"}, "title": "Auto-generated"},
)
design_id = r.json()["design"]["id"]
# Export as PNG
export_url = f"https://api.canva.com/rest/v1/designs/{design_id}/exports"
r = requests.post(export_url, headers=headers, json={"format": {"type": "png"}})
export_id = r.json()["job"]["id"]
# Wait for export to complete
# ... polling logic ...
return f"https://export.canva.com/.../{export_id}.png"
Pros: Designer-quality templates, easy for non-developers to create Cons: Requires Canva Pro ($13/mo), API is complex
Approach 3: HTML + CSS → Image (Puppeteer/Playwright)
Design your template in HTML/CSS, then render to image:
from playwright.sync_api import sync_playwright
def html_to_image(html_content, output_path, width=1080, height=1080):
with sync_playwright() as p:
browser = p.chromium.launch()
page = browser.new_page(viewport={"width": width, "height": height})
page.set_content(html_content)
page.screenshot(path=output_path, full_page=True)
browser.close()
# Template
template = """
<!DOCTYPE html>
<html>
<head>
<style>
body { margin: 0; width: 1080px; height: 1080px; background: linear-gradient(135deg, #1e293b, #334155);
display: flex; flex-direction: column; justify-content: center; align-items: center;
font-family: 'Inter', sans-serif; color: white; }
h1 { font-size: 64px; font-weight: 800; margin: 0; text-align: center; }
p { font-size: 32px; color: #94a3b8; margin-top: 20px; }
.logo { position: absolute; top: 40px; left: 40px; width: 80px; }
</style>
</head>
<body>
<img src="logo.png" class="logo">
<h1>{title}</h1>
<p>{subtitle}</p>
</body>
</html>
"""
titles = [("5 Python Tips", "For beginners"), ("AI Tools Roundup", "March 2026")]
for i, (title, subtitle) in enumerate(titles):
html = template.format(title=title, subtitle=subtitle)
html_to_image(html, f"post_{i+1}.png")
Pros: Full CSS control, designers can create templates Cons: Requires Playwright, slower than Pillow
Approach 4: Bulk generation from data
Generate images from a CSV/database:
import csv, os
from PIL import Image, ImageDraw, ImageFont
def generate_from_csv(csv_path, template_func, output_dir):
os.makedirs(output_dir, exist_ok=True)
with open(csv_path) as f:
reader = csv.DictReader(f)
for i, row in enumerate(reader):
img = template_func(row)
img.save(f"{output_dir}/post_{i+1}.png")
print(f"Generated post_{i+1}.png: {row['title']}")
# CSV: title,subtitle,category
# 5 Python Tips,Beginners,Python
# AI Tools Roundup,March 2026,AI
def my_template(row):
img = Image.new("RGB", (1080, 1080), "#1e293b")
draw = ImageDraw.Draw(img)
font = ImageFont.truetype("Poppins-Bold.ttf", 64)
draw.text((100, 400), row["title"], font=font, fill="white")
draw.text((100, 500), row["subtitle"], font=ImageFont.truetype("Poppins-Regular.ttf", 32), fill="#94a3b8")
return img
generate_from_csv("posts.csv", my_template, "output/")
Scheduling and automation
Run daily via cron
# Generate social images daily at 6 AM
0 6 * * * cd /path/to/project && python generate_images.py
GitHub Actions
name: Generate Social Images
on:
schedule:
- cron: '0 6 * * *'
workflow_dispatch:
jobs:
generate:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- run: pip install pillow
- run: python generate_images.py
- uses: actions/upload-artifact@v4
with:
name: social-images
path: output/
Common mistakes
- Inconsistent branding — use the same fonts, colors, and layout every time
- No template system — if you’re designing each image manually, you’re doing it wrong
- Wrong dimensions — Instagram: 1080x1080 (square) or 1080x1350 (portrait). Twitter: 1200x675. LinkedIn: 1200x627.
- Text too small — mobile users can’t read tiny text. Minimum 32px for body, 64px for titles.
- No logo/watermark — always include your brand. People share images without context.
- Not testing — generate 3-5 images and check them on your phone before automating.
Key Concepts
- Template = reusable design with placeholders for text/images
- Pillow = Python image processing library
- Playwright = browser automation (can render HTML to image)
- Bulk generation = creating many images from data (CSV, DB)
- Aspect ratio = image dimensions (1:1 square, 4:5 portrait, 16:9 landscape)
What to Try Next
- Automate social media posting with scheduling
- Generate QR codes for review links
- Build a content calendar in Google Sheets
- Create social media graphics with Canva templates
FAQ
Pillow vs Canva API vs Playwright — which to use?
Pillow = free, fast, code-based (good for developers). Canva API = designer-quality templates (good for non-developers). Playwright = full CSS control (good for complex designs). Start with Pillow for simplicity.
Can I use custom fonts?
Yes — download .ttf/.otf files and use ImageFont.truetype("font.ttf", size). Google Fonts has free commercial-use fonts. Store them in your project directory.
How do I handle different platforms (Instagram, Twitter, LinkedIn)?
Create separate templates for each platform with the correct dimensions. Or use one template and crop/resize for different platforms. Instagram: 1080x1080. Twitter: 1200x675. LinkedIn: 1200x627.
Can I automate video creation too?
Yes — use moviepy (Python) or ffmpeg to generate videos from images + audio. But video automation is more complex. Start with images; add video later. See Best Tools for Freelancers.