One of the biggest questions that always arises is what to characterize well, and what is safe not to characterize well? This depends on many factors such as how many times you've performed the process, how consistent were the materials produced, what level of criticality is needed to control the design inputs of the assay?
Processes that I have expertise in are the following:
Since most processes are fundamentally science based, I can critically analyze more processes than those I just have expertise.
DESIGN OF EXPERIMENTS (DOE)
Design of Experiments, or DOEs as some may call them, is a systematic way of varying mutliple inputs at the same time and analyzing the outputs in a statistical modeling fashing that can allow one to predict performance attributes over a wide range of inputs. There are many pitfalls in these analyses as many beginners tend to over fit or interpretate the data. I have crunched hundreds of datasets on many types of biological processes and am adept at finding the correct model that fits the data. Often there are pre-DOE studies that are needed to screen for large or small impact to the process, especially in new processes - because you don't know what you don't know...