Oh, I see! Well. That's a lot. So, your agents are posting the messages into a channel to ask questions, and then you want to gather the answers given -- but you also want to sort of build a little knowledge base and see if you can decrease the same questions being asked? Am I on the right track?
So I guess step 1 is to do the piece from the previous answer (so at least it was useful!) Get those replies for each of your messages. Do a loop to get all messages, and then do a loop within that to get all your replies.)
Next would be to have an Excel table ready to go, that you can store this information into. (Or a dataverse table, whatever works for you.) In your loop of list replies you would be adding into it.
Looking at it, the List Replies API gives us some of that data you'll need - the reply itself, of course, the created date time - and you'll have the original message too so you'll be able to use a Date function to minus that date from the message date, that'll give you the response time.
Now........... finding commonalities between the questions.......... hmm. HMMM. Hmm. Well.
Okay. There are .. options.
One would be to use AI. Get all your data crammed into that sheet / table, and then use the "extract most relevant words and phrases from text and then classify them into predefined categories". You'd need to do a manual pass there to define the categories, but that's going to make it much easier to start grouping them.
I think that's your best option with using the tools available. Is AI an option for you? (Before you answer, if you wonder why I'm asking that, take a look at this - https://powerapps.microsoft.com/en-ca/ai-builder-calculator/ )
