Fortune 500 companies can expect to push more than a gigabyte a day in raw web analytics data, which can be easily tripled for media companies. Big data is anything anyone ever talks about anymore, so the C-suite has never been more interested in integrated analytics, shining a spotlight on the web analyst team to deliver more than just pretty charts and high-level talking points.
Pulling the information from web analytics software should be less than 10 percent of the work, with an overwhelming 90 percent of time dedicated to deriving insights your organization can use to drive change.
So how to you go from pulling numbers to authoring insights?
1. Compare Trends, Not Just Differences
Web analytics software makes it extremely easy to compare equal periods of adjacent data, such as month-over-month or year-over-year, but other logical comparisons such as average weekday, current day versus the same day last week and other options are much more difficult to configure.
Unfortunately, the best way to find meaning in trends is by exporting data into Excel and crunching these numbers manually or by using pivot tables. You can then add layers of additional analysis such as calculating the long term mean, variance, and standard deviation.
2. Analyze the Significance of Your Data Before Drawing Conclusions
Nothing is worse than a web analyst that “cries wolf” over every little hiccup in a conversion rate. I once had a colleague that was very worried about a campaign’s performance, which dropped off sharply 8 weeks after launch, only later to learn that it was a back-to-school campaign and we were approaching Thanksgiving.
As discussed in the previous tip, calculating standard deviation is an easy way to determine whether the change you see in absolute numbers is statistically significant, if your data falls outside of two standard deviations of the mean.
3. Dig Deeper With Segmentation
Deciding on a driving force for statistically significant change is where you’re likely to spend 90 percent of your time in formulating insights.
Sometimes the driving force behind observed changes can be painfully obvious, such as broken functionality on a website, but other times a change can be like searching for a needle in a haystack. By segmenting your analytics data, you can quickly find commonly-shared behavioural traits that are influencing the changes in trends observed.
4. Correlate Reported Trends With Business Impact
This is the part of the report that should answer: why do I care? As a simple rule of thumb, try to attribute fair assumptions in revenue generation, cost savings, or visitor satisfaction back to the trends you observe.
For instance, did the landing page for a seasonal campaign perform significantly better last year? If so, how quickly could a change be made and what is the overall effect the change would make on bottom-line sales dollars?
5. Make Insights Actionable
The easiest way to make insights actionable is to derive ideas for a complementary optimization program. While there are many places to start optimizing, the goal for your web analytics reports is to include insights that can actually be completed within a short amount of time and have a significant impact.
It neither make sense to test modifications to pages with less than 1 percent of your overall site’s traffic, nor does it make any sense to recommend changes to pages beyond your organization’s control.
How do you make your web analytics reporting more insightful? Share your comments and ideas below!