How much metrics to follow up remains a challenging question. Ideally we would have a way at our disposal to reduce all our data to what is really needed. We might remove metrics that correlate a lot or metrics that do not show any changes over time.
Big data and an ever increasing volume of social media variables will finally put us to the limits of storage and processing. We cannot keep on storing all this data in the long run and need (analytical) methods to compress the information into more relevant information.
Principal Component Analysis is one of the relevant multivariate data reduction methods and insights that is fit for this. Using Analytical Intelligence to avoid the proliferation of reports furthermore should help focus on what really matters.