Hmmmm.....your question is a bit more complicated than you might think. Let me explain:
Average Minutes to First Reply, our best practice is indeed <1200, and the average that we saw across all of our communities (in August 2017, when this was last measured, which was a long process) was 1791 minutes. So this data is pretty clean and simple to understand. I could probably expound upon a few things that I do take umbrage with around the nature of this metric, but I'll save that for another time. 🙂
'Topics with First Replies' is not really a metric I understand, but I think I understand the spirit of the metric. I think it means that 75% of the Topics have at least one Reply, or conversely, that 25% of the topics have no Reply. We have historically had the metric, "Threads with no reply", which can be a bit of a maddening metric for a number of reasons. Not the least of which is that there will almost always be at least one Topic on a Community with zero Replies, or to be even more pedantic, there will almost always be at least one Topic on a Community that does not have a Reply during the time in which the measurement exercise takes place. Our best practice is that 70% of the Topics at any given time have at least one Reply. The reality is that (again, measured in August of 2017) that our communities averaged 41%. Ugh. Sad but true.
I like that you have a big fat red question mark with representing the Topics to Accepted Solutions ratio. We have not always been as good with quantifying that best practice as we could be. Conventional wisdom has been that a successful support-oriented Community should have a 'Resolution Rate' of between 20%-40%, but 'Resolution Rate' does not always nicely and mathematically correlate to the Accepted Solutions to Topics ratio.
Inevitably when you try to measure this ratio on a mass-level, you run up against the fact that not every Board and Topic (even for Support-oriented Communities) within our customer portfolio is looking for an Accepted Solution. So instead of trying to track down this ratio across are customers or even put our foot down and say it should be XYZ%, we throw our lot in with the study I cited earlier (which is platform-agnostic, btw) and say that the Resolution Rate should be between 20% - 40%. How do you measure Resolution Rate? You use our Value Analytics survey or run your own that asks the question, 'Did you find the answer you were looking for'.
Average Minutes to Accepted Solution is a wonderful metric because it can certainly be used to contrast the classic SLA that a brand might offer their customers. You can find how to get yours here. Sadly, Khoros has not run a benchmark on this (that I know of). We probably should though. Although I imagine that if the 'Minutes to First Reply' metric gets a little murky in terms of how to tightly quantify it to everybody's liking then 'Minutes to first Solution' will even more of a roller-coaster ride (what happens when there are two Solutions in a Topic and one is substantially better than another, or what happens when a Community has lots of great answers but is poor at marking them as such?).
Hope all of this helps.
