The numbers so far show that the difference between no compression and Gzip are vast, whereas the difference between Gzip and Brotli are far more modest. This suggests that while the nothing to Gzip gains will be noticeable, the upgrade from Gzip to Brotli might perhaps be less impressive.
The rub?
Gzip made files 72% smaller than not compressing them at all, but Brotli only saved us an additional 5.7% over that. In terms of FCP, Gzip gave us a 23% improvement when compared to using nothing at all, but Brotli only gained us an extra 3.5% on top of that.
So Brotli is just like spicy gzip.
Still, I’ll take a handful of points by flipping a switch in Cloudflare.
Worth noting that brotli level 1 is noticeably faster than gzip level 1, and is more or less equivalent in size to gzip level 6 for English text content, making level 1 brotli an effective choice for dynamic content; while level 11 is around 10% the speed of gzip level 9; but around 3–10% more savings; smaller than zopfli (which is around 10% again of brotli speed); making brotli great for pre-compressing static data to serve alongside dynamic data at a lower compression ratio.
In short:
you don’t require cloudflare to compress your data, you can generally do it as well if not better (especially in a primarily static application)
compress to the necessary degree for the resource
abuse technology like h2 and h3 to speed up ttfb (of the document itself)
pre-compress static resources using a tool like https://lib.rs/crates/static-compress at max power
abuse binary file inclusion/loading where possible