One of the most common issues in web site meaurement is having the data but not being sure what it means. Knowing whether the news is "good" is central to your company's success with web measurement.
When reviewing metrics, how do you know if "more" is better or worse? It depends on the metric you are looking at and the intent of your site. There are really two aspects of knowing if the news is good. One is understanding the performance of specific metrics: are more page views better or worse? The second is understanding how those metrics compare to those of competitors or other sites driving similar behaviors.
A few metrics are obvious: more sales or leads and a higher average sales price are generally positive (assuming product returns don't increase at the same rate as sales). Other metrics are more challenging. Examples are:
Page views per visit
Average visit length
Visits per visitors during a given timeframe
It's natural to assume that it is a good sign if all three of these metrics increase. But this isn't always the case. It all depends on the goals of the site. If you are running a content site supported by advertising, then yes, you would want to increase all three of those metrics. But if you are managing a lead-generation or online retail site, running these numbers up may not be a positive. For example, optimizing a lead-generation site may increase the number and quality of leads by 50 percent, yet reduce other important KPIs: