I logged what actually happened a while back, and it was more revealing than I expected. Looking back, how much of your social graph it wants is easy to check and almost never checked. The same thing happened reliably even when I varied the time of day. Whether it generalises I do not know honestly, rather than taking my word for it.

I approached this the boring way before deciding, and a few things became obvious quickly. Consistently, moderation quality predicts your reply rate better than anything else. That difference held up quite plainly on every account I set up. Worth testing yourself genuinely, before changing anything else.

If I were starting again, I would look at:

  • Look at whether the free tier caps daily messages — worth asking about before you invest time.
  • Ask whether you can undo an accidental swipe — worth two minutes before you commit.
  • Whether conversations sync across devices

I have been through this properly over a few months, and the results surprised me. The thing nobody mentions is that how much of your social graph it wants is the difference between two opposite reviews. It was measurable reliably for both of the accounts I tested. Worth a fortnight of your time frankly, after a fair bit of trial and error.