old Reddit actually has a worse time-to-interactive score It's an interesting example of a good metric that is perhaps not the right metric for a particular ap

old Reddit actually has a worse time-to-interactive score It's an interesting example of a good metric that is perhaps not the right metric for a particular application. For many/most users, I'm sure Reddit usage is something like 95% reading / 5% "interaction", or perhaps even closer to a 100/0 split. So "time until I can actually read what I clicked on" is probably the right primary user satisfaction metric...