Know your KPI’s before you chose web analytic tools

Avinash Kaushik tells us in his book Web Analytics 2.0 that often analysts do not get the respect that they deserve because they do not adequately measure one golden concept: outcomes (Kaushik, 2012). In fact, he believes that many of us spend so much time in the “weeds” reporting on detail data such as visits and time on site that we often forget to actually analyze the data and see what it means (Kaushik, 2012).

What is Outcomes analysis? Outcome analysis asks the question “Is my website getting it done?” “It,” of course, is the goal of your website’s existence (Herr, 2012).

This can create a real challenge for the web analyst. Since there are so many social media platforms available to the marketer, can a web analyst truly rely on one primary platform to provide the outcome metrics they need, or do they have to mix and match platforms based on the web media channels they are using?  Ultimately, the tools an analyst needs is going to depend in large part on the unique strategies and goals of their business.

In order to decide whether to use one or many platforms, we should revisit the concept of outcomes. If the ultimate purpose of analytics is to inform the business and to provide meaningful direction based on outcomes, then first the analyst needs to understand what is key to the business and what KPI’s need to be measured. Using these KPI’s, the analyst can then work backwards to which tools and platforms are going to provide the most meaningful data to analyze. For example, Chris Lake from eConsultancy states that “What you're ultimately looking for is a wide range of tools to help people interact. It doesn’t matter whether this interaction happens onsite or offsite, but only that it happens. You can measure it either way” (Lake, 2009).

Chris believes that the web is all about engagement, and the goal of any social engagement strategy needs to provide the right tools that allow people to make that engagement happen with a brand. Catherine Novak agrees with him and says that “Conversation is king, Content is just something to talk about” and conversation puts “human interaction at the centre of the picture” (Novak, 2010).  It is conversation that explains the rise of social media on the web and the growth of multi-user games on all platforms. To Novak, content without conversation is just broadcasting, or just advertising (Novak, 2010).

Chris Lake takes it one step further and provides a list of 35 social interaction metrics and KPI’s that an analyst can use to inform the kinds of tools and platforms that they need to have in place. A complete list of these KPI’s can be found at: http://econsultancy.com/blog/4887-35-social-media-kpis-to-help-measure-engagement but include the following;

1.   Comments

2.   Downloads

3.   Email subscriptions

4.   Fans (become a fan of something / someone)

5.   Favourites (add an item to favourites)

6.   Feedback (via the site) 

7.   Followers (follow something / someone)

8.   Forward to a friend

So, if a company’s outcome measures and KPI’s include engagement, then there are specific tools and platforms that will be essential to use for capturing metrics and ultimately analyzing these outcomes. Which social platforms are best suited to their needs and the resources they allocate will depend on knowing the ultimate outcomes that they are trying to measure but should include reporting on behavior and experience. According to an article in Communications Network, Herr explains this further:

Behavior - Behavior reports ask the question “What can we infer about visitors based on data yielded during their visits?”

Experience - Experience analytics ask the question “Why are visitors behaving in this way”

Behavior and experience data can come from clickstream reporting that is available through free tools such as Google Analytics or Yahoo! Web Analytics. However; for this general reporting to be most useful, Kaushik suggests drilling deep into reports and, for example, analyzing site search keywords to determine what visitors look for when they visit, or viewing site overlays for a better understanding of how visitors navigate the site.

In his article, Herr provides an important side note that clickstream data can only generate inferences, and different inferences can be drawn from the same data set. This is important to consider when reviewing this data with your team (Herr, 2012). Clickstream data can be considered part of traditional analytic tools and will only take you so far. If a company’s KPI’s include behavior and experience on their website for example, they may want to combine a traditional analytic tool with tools from Tealeaf, Clicktale, or RobotReplay that record all sessions on the website and can provide video playback of the entire customer experience (Kaushik, p. 136).  These specific tools now allow the company to combine baseline data with super rich data for a deeper picture into the behavior and experience of their customers as they engage with the website.  

There is also a school of thought that is more focused on conversions, goals, and Return on Investment. Professionals in this camp have a different set of outcome measures that matter to them. Kaushik has many excellent ideas for companies that are tracking these KPI’s. For example, he states that you can analyze a lot of ecommerce outcomes by measuring organic search. Although organic search makes up a small percent of a website’s overall traffic, Kaushik states that it accounts for a much larger contribution to multiple conversions (Kaushik, p. 108). If a company’s primary KPI’s were focused on conversions on their ecommerce site, then they could continue to use the same traditional analysis tools of Google Analytics or Yahoo!, but then combine them with more robust SEO tools to capture and analyze revenue, average order, products sold and bounce rates. These additional metrics will allow the company to see if they have lost sales opportunities, or if they are maximizing the revenue potential of search for ecommerce.

Bottom Line? While there appear to be many platforms that a web analyst can implement to help create the most robust toolset, the best place to start is with Google Analytics or Yahoo! Web Analytics which provide an excellent baseline regardless of whether your outcome KPI’s are based on transactions or engagement.

From there, an analyst should determine the top set of management and business KPI’s and overlay unique tools to capture the metrics necessary to gain deeper and wider picture of these specific behaviors and outcomes.

It is easy to get caught up in the myriad of tools available today, so take this last word of caution and implement the Kaushik 90/10 Rule: Spend 10 percent of your web evaluation budget on tools and 90 percent of it on people. Deriving meaningful insights from the mass of reporting noise requires a skilled mind, so make sure you find one (or cultivate one). (Herr, 2012)

Want to learn more? Visit these sites:

Herr, L. (2012, September 11). Moving from Web Analytics to Web Evaluation. Retrieved January 26, 2014 from: http://www.comnetwork.org/2012/09/moving-from-web-analytics-to-web-evaluation/

Kaushik, A. (2010) Web Analytics 2.0. Wiley Publishing, Inc. Indianapolis, Indiana

Lake, C. (2009, October 30). 35 social media KPIs to help measure engagement. Retrieved January 21, 2014 from: http://econsultancy.com/blog/4887-35-social-media-kpis-to-help-measure-engagement

Novak, C. (2010, July 27). Why conversation, not content, is king. Retrieved January 21, 2014 from: http://socialmediatoday.com/wordspring/152636/why-conversation-not-content-king