In my role as a board member of my local chapter of the Association for Women in Communications, I’ve been compiling analytics. I collect the information through Google Analytics (website), MailChimp (e-newsletters), and Hootsuite (Twitter), and then I give a brief report at our monthly meetings. Sounds simple enough, right?
Not so fast. Not for me, anyway.
These tools can be really useful for communications, but every time I log in for analytics, I feel like I’m bushwhacking with no machete. The statistics are piled high, and the graphics look appealing. But every month, I log in, look around, get overwhelmed, consider downloading default data, decide I can’t really use that, get mesmerized by the colorful charts, and start clicking around randomly like a lab monkey in a cage. After wasting an hour or two, I log out and take a nap.
Last month, I determined to have a strategy before I logged back in.
My first step was to think about how I would use the data; imagining myself giving the report to the rest of the board helped me focus:
- The report had to be brief
- The data had to be explainable to others
- The information had to be meaningful, and therefore specific and applicable to our group’s overall goals
I like to procrastinate think about the big picture, so next I rewatched a couple of insightful talks on information glut:
Clay Shirky’s “It’s Not Information Overload, It’s Filter Failure.”
What I like about Clay Shirky’s talk is that he puts our current situation (his talk is from 2008) in historical context. We’ve “evolved” in an environment where publishing and printing have traditionally had high upfront costs, and therefore publishers acted as gatekeepers and quality-control filters. Now we have low-cost, low-barrier-to-entry publishing and we have spam, “information overload,” a flood of low-quality content, and a whole slew of new privacy-control issues.
Our new forms of media require new ways of filtering because we no longer have a filter (via cost and inconvenience) near the source of information. And because we depend on email, social media, search engines, blogs, and many other new means of getting information, we need new ways of thinking about filtering. The information sources we use now are often not in “linear” form—a filter mid-stream will not work if the stream is actually a connected group of lakes.
Which brings me to JP Rangaswami’s TED talk, “Information is Food.”
Rangaswami is on to something—on to a new way of thinking about filtering—when he says that information is food. At the end of the talk he asks:
If you began to think of all the information that you consume the way you think of food, what would you do differently?
Analytics, I thought, is an all-you-can-eat buffet. How to avoid indigestion at a buffet? Be choosy: go for quality and freshness, don’t mix too many types of foods, and don’t be tempted to overload your plate just because everything is so cheap.
Now it was time to devise my plan for going into the jungle of data, and coming out with something … digestible. I came up with a few simple questions that I thought our analytics could shed light on, and I put them in three rough categories:
What people like
- What stories get the most click-throughs in the newsletters?
- What links or tweets get the most clicks/retweets?
- What subject lines result in the highest open rate for newsletters?
Experiments and outliers
- Did we change our methods this month? Can we see results in the analytics data?
- Are there any unusual results (compared to industry benchmarks, for example)—either very high or very low?
What brings people in
- What search terms were used to get to the website?
- What search terms were used to search within the website?
- Where are people being referred from?
There are many other good questions one could ask, and there’s an endless amount of minutia one could track. I chose these because they seem to suit our needs. And I like to group them in categories because it makes it easy to pick a question or two from each category without feeling like I need all of the questions every time. The important thing, I now believe, is not to try to consume more information than I’ll actually use.