One thing to keep in mind when digging into your analytics data is whether or not your research has the possibility of causing you to do something. Many pundits like to refer to this as making it actionable. I think this is a pretty good way of expressing it.
I happen to prefer a slightly different phraseology. Will it change what you are doing today? If you found out that your bounce rate was up by 10%, would you act on that? If your return on ad spend (ROAS) in your PPC campaign drops by 25% will you act on that? If the website that is referring you traffic suddenly slows to a trickle, would you act on that? Probably for all 3 of these examples.
If you found out that your blue widgets sold better in Virginia than Florida, would you act on that? Maybe you would, but it will depend a bit on your site and business structure. You may not be able to act on such data. Businesses with more robust web sites and capabilities would love to have that kind of data. Businesses without the ability to react to such data won’t care.
This is one reason why analytics is so individual to each given business. One of the most difficult things to do in analytics is to figure out what data matters to you, and why. It’s a complex process, and you best be prepared for lots of experimentation to get the best results.
