Planning, conducting, and analyzing experiments performed in dynamic processes, such as continuous processes, highlight issues that the experimenter needs to consider, for example, process dynamics (inertia) and the multitude of responses. Dynamic systems exhibit a delay (transition time) the change of an experimental factor and when the response is affected. The transition time affects the required length of each experimental run in dynamic processes and long transition times may call for restrictions of the randomization of runs. By contrast, in many processes in parts production this change is almost immediate. Knowledge about the transition time helps the experimenter to avoid experimental runs that are either too short for a new steady-state to be reached, and thus incorrect estimation of treatment effects, or unnecessarily long and costly. Furthermore, knowing the transition time is important during analysis of the experiment.Determining the transition time in a dynamic process can be difficult since the processes often are heavily instrumented with a multitude of responses. The process responses are typically correlated and react to the same underlying events. Hence, multivariate statistical tools such as principal component analysis (PCA) are often beneficial during analysis. Furthermore, the responses are often highly positively autocorrelated due to frequent sampling. We propose a method to determine the transition time between experimental runs in a dynamic process. We use PCA to summarize the systematic variation in a multivariate response space. The time series analysis techniques ‘transfer function-noise modeling' or ‘intervention analysis' are then used to model the dynamic relation between an input time series event and output time series response using the principal component scores. We illustrate the method by estimating the transition time for treatment changes in an experimental blast furnace. This knowledge provides valuable input to the planning and analysis phase of the experiments in the process.