There are limitations with Splunk not detecting all user activity, especially on mainframes and network devices. This is because Splunk relies on agents, which cannot access certain workstations. In these cases, we have to rely on application data. For example, with mainframes, manual reports are generated and sent to Splunk, limiting visibility to what's manually reported. This lack of automation for specific platforms needs improvement from Splunk. Additionally, API access is limited for other applications that rely on API calls and requests. This requires heavy customization on Splunk's end. These are the main challenges we've encountered. Monitoring multiple cloud platforms, like Azure, GCP, and AWS, with Splunk Enterprise Security presents some challenges. While Splunk provides different connectors for each provider, consolidating data from two domains across distinct cloud environments can be complex. However, leveraging pre-built templates and Splunk's data collation capabilities can help overcome these hurdles. Despite initial difficulties, I believe Splunk can effectively address this task, earning it an eight out of ten rating for its multi-cloud monitoring capabilities. While Splunk Enterprise Security offers insider threat detection capabilities, its effectiveness could be enhanced by integrating with additional tools, such as endpoint security solutions. This integrated approach is particularly crucial for financial institutions, which often require dedicated endpoint security teams. While using multiple tools is valuable, further improvements within Splunk itself are also necessary. Considering both external integration and internal development, I would rate its current insider threat detection capabilities as three out of ten. Threat detection is where Splunk falls behind. While it offers tools, other use cases require additional work. PAM is an enterprise tool that centralizes information about users, servers, and everything else. It needs real-time monitoring, which I haven't seen in any of the companies I've worked for. They only rely on Splunk for alerting, but real-time monitoring should be handled by the endpoint security team's tools. This means there's no detection or analysis at the machine or endpoint level. Additionally, threat analysis reporting is also absent.
