Often considered hopelessly out of the money in the horse race of web analytics, governments around the world still publish guidelines and establish standards, all of which are available for free. Plus, the fact they are not necessarily cutting edge makes them good yardsticks by which to measure where the broad middle settles when it comes to standards, understanding, and expectations as regards to analytics.
Regrettably this writer is proficient in just the one language in use herein, and so my survey will encompass only the content of the U.S., U.K., Australian, and New Zealand government sites. I think of it as a small compensation that I can, in the case of New Zealand, think of web analytics and “The Lord of the Rings” at the same time (which is chiefly why I chose it over Canada).
United States: “Helping Agencies Deliver a Great Customer Experience”
This site takes users from an explanation of “what is website evaluation/web metrics?” all the way to “tying website evaluation to mission achievement.” There’s also a significant section of “quality and compliance.”
According to the site, developing a web metrics strategy involves collecting “only the data you need”; identifying “action areas” for improvement; determination of benchmarks; and determination of ROI.
The page itself was updated earlier this year. However, the “sample framework for measurement” (downloadable) is older than Methuselah in Internet years (2005).
The site is fairly straightforward, in typical U.S. fashion. In my brief review, I could find little about “how” to accomplish all of this, and one assumes the government would readily acknowledge that expertise from outside the agency will be needed to accomplish accurate, reliable reports no matter the tool in use.
According to this site, every publicly accessible site (associated with the government) must comply with standards developed by The Joint Industry Committee for Web Standards by tracking at least as follows: unique browsers; page impressions; visits; visit duration. There are several other requirements that must be reported back to the Central Office of Information (COI) – most are quite basic.
When the site gets around to “using web metrics for planning and design,” it suggests that “the analysis, interpretation and combination of web metrics in order to understand and optimize website usage is known as ‘web analytics.'” Huzzah!
The site goes on to cite more advanced metrics for the purpose of planning and design, but I’m not sure I would agree with the emphasis. For instance, it says much about internal search, which I agree is important but not the next thing in importance after unique visitors (using their priorities). Under “user experience,” they do cite “tracking a user’s journey,” a worthy cause, as follows: the search engine or website that drove the user; entry and exit pages; route taken by user through site. “Referrals In” is mentioned as a way of understanding user interests; and there is mention of the usefulness of qualitative data as well. It even lets “expert reviews” into the sandbox (also known as “heuristics” or in my opinion, “opinion”).
The government of the Australian state of Victoria relies on the Web Analytics Association’s (WAA) definition of web analytics, which is: “Web Analytics is the measurement, collection, analysis and reporting of Internet data for the purposes of understanding and optimizing Web usage.”
Victoria has already standardized on Google Analytics, so it has cut short the discussion of tools (though it does seem to allow agencies to make a case for using other tools if they wish). Beyond this, it cites the following rudimentary guidelines to track:
- Popular content
- How users navigate their website
- The search engines and keywords used to access its website
- The geographic location of its users
- Browsers and operating systems used
- The effectiveness of marketing campaigns
One interesting note about the site is that it has its own version of a bit.ly-like URL shortener; surely a convenience that even a hidebound person in the U.S. might want to spring for. The site also notes that a “supplier must be either free or near-free or must go through typical government procurement process.” Fun!
New Zealand: “Web Analytics for the New Zealand Government”
The New Zealand site strikes me as the most comprehensive of the four. Going well-beyond rote rules and descriptions, New Zealand provides matrices for understanding log file vs. tag-based solutions, lengthy descriptions of the terms, and concepts often associated with best practices – including a fairly robust definition of what a “KPI” is, including the following example of KPIs in the real world: “KPIs should answer the question ‘why is it that we spend all this money on website.govt.nz?'”; and indeed, this is the question KPIs must answer for any organization. A deceptively simple concept, we know it has eluded many in the marketplace, but apparently not the folks in charge of Kiwi analytics.
The site even includes a matrix for what types of information are important to track for different types of sites. This is similar to the process I’ve earlier described in the “e5o” process with its companion five site-types and stages of customer engagement.
Demonstrating its currency, the site also has a section on social media, but we have no space for analysis of that in this column.
The New Zealand government has taken a great deal of care to be both clear and comprehensive on this site, so if anyone is looking for a primer that’s been provided by a large, official source, this may be the place to go looking. And of course, remember that as you track your KPIs, there is, somewhere in New Zealand, just One Ring to Rule Them All.
I must make the disclaimer that I haven’t researched any of the four sample governments’ privacy policies, as that would require more space than we have here.
In summary, I hope this column has done at least a couple of things: provided easy access to some valuable resources about web analytics; and put into perspective the way governments, roughly equivalent to a large part of the bell curve, stand in relation to the maturity of web analytics efforts.