Defining Web Analytics
Web analytics must be about more than site performance.
Web analytics must be about more than site performance.
One thing worries me about Web analytics: the name. “Web analytics” sounds a bit scary, doesn’t it? Those two words don’t necessarily sit comfortably with your average marketer or business leader. When somebody asks you what you do for a living, “Web analytics” doesn’t tell them much.
For a long time, nobody really knew what it meant. There were heated forum debates about what Web analytics was and wasn’t. Some people asserted Web analytics was essentially about what came out of a Web analytics tool; others believed it a much broader discipline than that.
So, it was good to see the Web Analytics Association (WAA) recently publish a draft definition of Web analytics:
Web Analytics is the objective tracking, collection, measurement, reporting and analysis of quantitative Internet data to optimize websites and web marketing initiatives.
It’s great to have a definition, but does it help the average business person understand what it is and what it does? Jim Sterne, president of the WAA, acknowledges it’s a start, and you have to start somewhere.
Web analytics is very focused on the site, not the people who use the site. It’s a bit like the old days in consumer packaged goods, when brands focused on selling products rather than marketing to people. We must think about optimizing Web site performance from a much wider perspective.
An effective strategy for measuring and optimizing Web site performance has four key components:
Market intelligence provides the context for a business’ performance. Though the majority of a digital marketer’s time can be spent focused on the brand and site, neither the brand nor the site operate in a vacuum. External factors and forces are at play, including:
These and other factors are likely to influence the way the marketing plan is developed and executed. For example, are your growth rate assumptions in line with overall market performance? What would the effect of rising media costs be on the expected level of return on investment (ROI)? What competitive scenarios might affect your ability to meet your targets? This type of market intelligence data often comes from third-party data sources, such as panels (Nielsen//NetRatings, comScore), aggregators (Hitwise), and large-scale surveys (Forrester’s Technographics).
Data on users usually comes from users. This includes the use of surveys, observation, and focus groups as well as looking at behaviors on the site.
User profiling is the process of getting to know who uses your site and why. Basic marketing principles are about understanding customers and meeting their needs. It’s no different on the Web. By getting to know and understand your users, their behavior, needs, and desires, you’ll be in a much better position to:
Finally, site performance measurement looks at a site’s effectiveness from a technical perspective. It concerns such site aspects as the page-delivery speed, site availability, and responsiveness of transactional processes.
Site performance measurement is often neglected when we think about measuring channel effectiveness. It’s often seen as a technical issue rather than a business one. It’s the preserve of the operations team rather than the marketing team. However, in terms of understanding how well the marketing is performing and how it can be improved, you must do that in the context of how the site is performing as well.
If panel data and survey analysis and interpretation aren’t part of Web analytics, for example, what is it? It’s part of a wider e-marketing discipline. It’s more holistic and puts the focus on the consumer rather than the site or channel. Increasingly we talk about “e-CI,” or e-business consumer insight. It’s an approach that puts the consumer at the heart of the matter.
Neil is off this week. Today’s column ran earlier on ClickZ.