Right, so my particular use-case is scraping SPA's for meta since they don't initially have it on load. Because this can be slow (thanks a lot JavaScript), I needed a way to parallelize multiple instances so my app remains snappy. This eventually drove me to write Navalia.

Other ones I can think of: - Testing (run a series of acceptance tests against a website) - Screenshots - Rendering of PDFs - Web crawling (especially for SPA's)