In talking to a few people about the photo time-lapse prototype I built (and in doing some simple math), I realized that it wasn't a great idea. Just creating a short 30-second time-lapse video would require uploading ~9 gigabytes of photos (30 fps * 30 seconds * ~10 MB/high-res photo).
Average user upload speeds and the immense server storage requirements alone would make it very challenging and unlikely to catch on.
But through this experiment I came up with the another idea: An automated website screenshot service. The similarities were in being able to see a time-lapse view of a website as it changes over time.
I started developing the app (React, Python/Flask, MongoDB) and took a day to design a simple logo, a landing page, email capture form, etc.