Great blog very useful, thank you! (As opposed to the learn.microsoft.com article which has quite a few mistakes in the PowerShell and some missing context)
I have been working with a client to set up this integration and we are the point now where it appears everything is working correctly and the interactions are viewable in Activity Explorer and searchable via eDiscovery, which is great.
One observation we've made is that the "chaining" of conversations or prompts and responses is not really built in or intuitive and interactions seem as though can appear out of order based on what happened within the minute, or within the synchronisation. This can also be the case when you export the interactions via eDiscovery search, meaning it takes additional effort to identify which response is for which prompt etc.
Is this a known thing, or have we missed something? If known is there any plan to improve the readability and chaining of interactions both in Activity Explorer and eDiscovery?