For data correlation, we just haven't worked long enough with it to assess its impact on our overall threat response strategy. We prefer a system that simply informs us when there is a problem; we don't want to engage too much in threat hunting. Therefore, we're not looking to create a SOC from this, which is also why we moved from XDR to MDR. There are areas for improvement, including the difficulty in getting the right handles on the applied email filters. It's sometimes unclear why one email is treated as spam and another is not, even if they contain similar content. Making the process of how emails are treated a bit more transparent would be beneficial.
