Determine Which Questions to Answer
Okay, we've all done it at one time or another. That doesn't make it any less nonsensical, though.
All reports in a web analysis answer a question. Sometimes the questions they answer are pointless. In fact, often the high level reports right out of the box are useless.
Think about it for a minute. What does the number of visits to a site tell you? Does it tell you if the website is effective? Is the marketing working? Are these worthwhile visitors? Is this improvement? What do we do next?
No, it just tells you approximately how many visits were made to a given site for that period of time. It tells you nothing about the quality of those visits. It tells you nothing about why those visitors came to the site, what they did, whether they were satisfied with the site, what their expectations were. Most importantly, it gives you no indication of what you should do now.
Even if you graph this number over time to illustrate a trend, the question you are answering is, "How many visits were made to this site?" Or perhaps, "Are there more or less visits on my site compared to the past?"
Why, then, is this the most commonly requested report? Why do the higher-ups all want to see this graphed out week after week? Unless you make money every time a visit is made to a site, this statistic doesn't do very much for you.
The reason this and any number of other canned reports are so frequently requested is because we tend to look at a report and then try to determine what it's saying. We find something interesting and then try to interpret.
This is backward. This is like finding an answer and then trying to figure out what question you were supposed to ask to get it. This strategy might work for a game show like Jeopardy, but it's a terrible way to run a business.
It demonstrates that we don't know how to use web analytics tools yet. With all the neat reports and graphs and colors, too often our heads spin so fast with all the new tools that we forget why we opened the program in the first place.
"What Do We Do Now?"
With all the things web analytics tools do for us, their main purpose is to help us figure out what to do next. The question we ought to be able to answer after reviewing reports is "What do I do now?" In a word, we need to make web analytics actionable!
We do this by starting with a list of questions we need answered. For example,
- Which marketing mediums are underperforming and need to be dropped?
- Which pages on our site are keeping visitors from accomplishing their goal?
- Which marketing messages seem to be giving us the most confused visitors?
- What are visitors' intentions when they come to the site?
The list can go on, and it will vary with each business. It will likely vary over time as well.
(If you've exhausted all those questions, start with Analytics Intelligence. It will start asking questions for you.)
The grand key, then, is to take these questions and find the reports that answer them. And until those questions are answered or you come up with more to ask, all the neat charts and graphs the analysis tool makes can be ignored.
Start Here with Analysis
These questions should be customer-centric as well as being oriented on the company's stated goals. Don't just ask whether visitors are buying, but also ask what other reasons bring visitors to the site. Don't just ask how high your conversion rate is, but also ask whether visitors have the same goals as you expect them to.