Hi Marisa!

Thanks for sharing your challenge with us here.

I actually tried previously to create some benchmark data on this. The problem really is that many communities do it different: The frequency, the wording, how many questions, but as well as to when this question is being asked… As everybody is doing it a bit different, it seems not really comparable.

I also see variance in the CSAT score itself: For some communities it is higher than average, while for some it is lower. I am not sure if this has to do with the channel itself (only offering support content or mixed content), or if this is being caused by other factors (e.g. how many Super users, Community team activity, etc.).

Mostly I also see people using a lickert scale from 0-10 to measure CSAT, however some like to branch it out a little, or incorporate this in their survey for support deflection. In this case, they are asking it something like this:

• How happy are you with your visit to the community? (1-10)
• Did you come to the community looking for an answer to a question? (Yes/no)
• If yes: Did you find the information you were looking for? (Yes/no)
• If yes: Would you have contacted support instead? (Either yes/no or a selection of channels)

What is really important is when you ask them about their satisfaction. I see sometimes that a survey pops up after 1-2 minutes, which is too early in my opinion. I think the survey should be displayed earliest after 4-5 minutes. When do you trigger the survey to be shown?