Taiichi Ohno preferred facts over data, meaning, among other things, that direct observation trumps second hand stuff. How else can you truly grasp the current situation and identify the waste?
Well, the fact of the matter is that direct observation is not always practical. Sometimes it needs to be supplemented with what I call forensic observation. For example, if we need to gain an understanding of the pre-kaizen situation for a REALLY long lead time process, say weeks or months or even years, it's not very pragmatic to grab some time observation forms, stopwatches, spaghetti charts, etc. and...camp out.
No. Long lead time processes, such as bodily injury insurance claim evaluation or complex business proposal development, often should be subjected to the rigor of process mapping. These process maps detail the historically and forensically based steps, hand-offs, rework, waiting, etc. as supported by emails, documents, system entries, recorded phone conversations and the like for specific, real-life claims, files, proposals, design projects and so on.
It's a bit like CSI, but without real blood. It's a pragmatic proxy for going to the gemba and it can be bolstered with true direct observation for specific steps within the process. For example, we can directly observe how design requirements are identified from the request for proposal (as part of the overall proposal generation process).
Now, this is not a license for dismissing direct observation (and it's not a replacement for appropriate value stream analysis). But, given the right circumstances, forensic observation can be an appropriate way to apply gemba-based principles to your kaizen activity.
So, how have you applied forensic observation within your lean journey?