How to Migrate Hundreds of Legacy Images to Your Own CDN
282 images on a third-party CDN. A matching algorithm, a side-by-side review tool, and a canvas-based format fixer. From liability to asset.
April 11, 2026
Why this matters
Every image in a course is either an asset or a liability.
Connected to the rest of your site, an image builds topical authority. Search engines see the link. LLMs see the relationship. Visitors discover related content. One image creates three connections.
Disconnected on a third-party CDN, an image decays. The host changes URLs. The format doesn't match your archive. Nobody discovers the 400 visuals behind the concept because there's no link between them.
282 images across 4 courses were hosted on Mighty Networks. The platform we migrated from. Zero connections to the visual archive. Zero internal links. Every image a dead end.
34 more lessons had no images at all. Text-only. Missing the visual reinforcement that makes VV courses work.
The scan
First pass: extract every image URL from every lesson's HTML in the database. Group by source.
| Course | Images | Source | |---|---|---| | Build Once, Sell Twice | 121 | Mighty Networks | | How to Visualize Value | 67 | Mighty Networks | | Permissionless Apprentice | 53 | Mighty Networks | | Fundamentals of Value | 42 | Mighty Networks | | Context is Capital | 17 | VV CDN |
Only Context is Capital — built natively on this platform — uses VV CDN. Everything else points to a server we don't control.