A while back I finished reading the book titled Web Analytics: An Hour a Day written by Avinash Kaushik. Without a large amount of awareness on web analytics training or prior experience in the web analytics field, it turns out the book and his principles are fairly well-known in the industry and among colleagues. The author has subsequently written a follow-up book titled Web Analytics 2.0 which is on my short list. Although the book is specifically geared towards web analytics terms and processes, some of his advice is applicable far beyond web data and syncs well with general marketing research and data analysis techniques.
Here are 6 of my largest takeaways from the book Web Analytics: An Hour a Day.
Analyze beyond the data
It's one thing to execute data reports, pull numbers, and put them into a fancy chart or graph but web analytics requires more than these basic duties. As I've mentioned before, the data is just the beginning of your process and you need to spend equal (if not more time) on analysis and reporting to apply insights.
Put yourself in your reader's shoes. Ask yourself why this statistic is 60%? Why has it increased from 40% over the past 3 months? What are your interpretations and assumptions? It's okay to be wrong, that's why they're called assumptions. Provide a dashboard but don't stop there, go the extra mile by summarizing themes and highlights. Most anyone can read a chart, but interpreting the chart is where web analytics professionals truly add value to their organization.
Linking the "what" to the "why" using Voice of Customer (VoC) and market research
It always comes back to the customer. In a world of big data and website analytics that is only growing large by the second it is easy to get overwhelmed and caught up in the "what?" The what is how many page views, how many visitors, average length of visit, etc. It's all of those metrics that you are pumping out through automated reports and sharing with staff.
Unfortunately spending all of this time on the "what" robs you of spending time on the "why." The why is what really matters. It's talking to your customers, readers, and visitors of your site. It provides context as to what your customers like, what they dislike, and answers questions like: if an average time on site is 30 seconds, is that is good or bad? The customer may find what he or she was looking for easily and they place a call or send an email, or they may find your site frustrating enough that they cannot last longer than 30 seconds.
The 10/90 rule.
Kaushik makes the point here that web analytics is more about the people than systems. Although it helps to find professionals that are experts in Omniture, Adobe, Google Analytics, etc. - most people can train on systems. What is more difficult to find is those who are strong analysts and strategy builders. Systems people are often more IT-oriented than data analysis-oriented. His 10/90 rule supports this saying that for every $10 you spend on web analytics systems, you should be investing $90 in intelligent resources and analysts.
The 3 key questions of website traffic: how many visitors, how long did they stay, how many pages did they visit.
He argues more so than anything else, these are the 3 metrics you should care most about. Visitors determine the reach of your website and how many people your engage with your brand. Length of time on the site indicates depth of engagement and whether or not they are interested in more content. Pages visited indicate a long-term commitment among visitors. Ideally you want visitors to enter your site for one thing but stay for others and build them into regular users. This varies by site purpose, but in general most would agree, growing new visitors, return visitors and engagement is key.
A fairly simple acronym here to keep in mind when working in web analytics: Define-Measure-Analyze-Improve-Control. Using this principle, it will help you walk through chart-by-chart in your report with the same focus in mind. It's a continuous improvement process and asking the right questions is arguably more important that always finding the answers. Questions generate good discussion and strategy.
Don't try to be a numbers god
One might think you want a person "who eats Excel spreadsheets for breakfast" as Kaushik would say, but this is not the case. You want to find a person that is more strategic. Focus on hiring talent that can both pull and analyze the data but also take the next steps and apply context and meaning to the numbers. The author states "numbers gods" often get trapped in web analytics because there is too much ambiguity as in 1 + 1 = 3 sometimes. Pure statistical gurus cannot come to grips with this. To be frank, I never will be able to either but I wouldn't consider myself a "numbers god."
The Drive Research blog from our market research company in Syracuse, NY is designed to provide helpful tips and content to those working in the market research and analytics field. Questions or comments about any of our posts? Give us a call at 315-303-2040 or email us at firstname.lastname@example.org.
Liked this blog post? Try reading these similar posts: