Google Analytics vs. WebTrends – the Government Debate

I’ve been asked a couple of times about whether I think ok for Departments to install the free version of Google Analytics on their sites or continue to mire through the internally-hosted, yet more complex to install & manage, WebTrends (or one of their competitors).

Before you read any further, I’ll come clean: I don’t know the answer, and I can’t decide this for you even if I had an opinion one way or the other.  But I will share with you some questions I am asking myself and my colleagues that may help inform conversations that may be going on in your department.  I beg you to share the results of your conversations as I’m desperate to know for my own department which way I should be pushing.

As you know, I’m a fan of free.  But is Google (anything) Free as in Freedom or just Free as in you don’t have to pay? And if it’s free for us, then what’s in it for them?  Don’t get me wrong, I love Google.  I’m still kicking myself for not buying shares when they went public.  I’m even using it to track the uptake of my RSS feed, but I’m not the government. However, in the growing debate over privacy and security issues on the web, we need to take caution in anything that could even be *perceived* as putting information-as-power in the hands of one US-based corporate for-profit entity. We have a responsibility to protect the security and rights of citizens no matter how small the risk.  I don’t want to promote alarmist views, but here is an interesting FICTIONAL sci-fi thriller what-if scenario to consider…

Alternatively, if Google had a tool that could be installed on internal servers to use (do they already? I know they do for their search engine) I imagine that WebTrends would be losing even more market share to them.  However, until we sort out the risk question of using the hosted-by-them version, perhaps we should do as one presenter suggested at the eMetrics conference last year in Toronto and “love the one we’re with”.

I’m not in favour of spending money on something when there is an excellent, free, alternative out there that is easy to install, use and interpret.  However, I do think we need to engage our IT security folks and do our homework before we do.  So if anyone has already done a risk assessment of Google Analytics, can you let me know or post it on GCpedia?

Regardless of what tool you are using, the more important focus should be on investing in people who know how to interpret the data, sharing it with decision-makers and make changes to improve the site based on what users – citizens – need.

About these ads

About Laura

As a Business Analyst working for the Canadian federal government in Web usability, I have the opportunity to be a part of a growing movement of professionals implementing user-centered design principles.
This entry was posted in Measuring success and tagged , , , , , , , . Bookmark the permalink.

19 Responses to Google Analytics vs. WebTrends – the Government Debate

  1. Rod Maides says:

    Good points. But I’d suggest that even before deciding what tool to use, readers need to ask the key question, ‘what’s the objective?’

    Many of us are still getting requests for “hits” as a measure of success rather than data that might show cause and effect in marketing a particular page, referrers or program or entry-exit data that will tell you how users got to the site, what pages they viewed and their exit point.

    For me, I was a long time user of NetTracker (eventually bought up by another firm) because it provided what I needed in graphics (for time pressured execs) as well as the granular data necessary to convince content developers of a particular step.

    Above all, being able to parse through log files in-house was a prime consideration. In that way, we were able to convince the powers that be that our data was indeed “ours” and not accessible to outside interests.

  2. Laura says:

    I agree completely, Rod. I’m also getting the not-so sophisticated requests for visits, page views, bounce rates and referrers, although there is value to be found, especially with some ratios and intelligent reporting. Sometimes even I find it tough to avoid those traps.

    But you’re totally right, we need to identify the goals that meet the requirements of the government (according to the policies) and make sure that citizens can do what they need to do.

    Actually, I’m just about to embark on a really neat project to do just that – identify tasks citizens said they wanted (in surveys and e-mails) and all the policies that we need to follow, organized by target audience segment so we can organize the site for the audience rather than by our organization structure. That would allow us to test success factors to challenging and validating our assumptions.

    At least, that’s the idea.

  3. Hey Laura, it so happens that I’m looking into Urchin, Google’s for-sale version, which Avinash Kaushik says is 0.5 versions behind the free one. If it’s as cheap as some sites are leading me to believe (less than $3k) and is easier to put on an internal server than WT, then there might be another tool for the paintbox (and more importantly, the data is housed within a department’s firewall). I find GA much easier to use for segmentation than WT. Best of all, I can configure the reports myself and not rely on IT for that service.

  4. Laura says:

    Nice one Denise! I look forward to hearing more!

  5. Alex says:

    eGov solutions should minimize risk and maximize functionality so a “behind the firewall” solution is required but consider the new ways to capture this data and leapfrogging the naive users who still ask for hits (How Idiots Track Success-HITS) to the next generation of tagless solutions like Atomic Labs Pion that takes the value into ANY target.

  6. Lateef says:

    Great conversation here – I checked out the Urchin demo and the GUI stinks compared to Google Analytics – leaving an obvious choice for WebTrends

    You can see a demo of Urchin here:
    Username: demo
    Password: demonstration

  7. Laura says:

    I’d like to do a part 2 to this post so if anyone has ideas, experiences or resources, feel free to add them.

  8. Aaron says:

    I think it’s worth bringing into play Google’s “new” opt-out plug-in and how that helps citizens concerned about privacy/tracking

  9. Hi Laura,
    I’ve done quite a bit of research on this, and even though Google does hold your data (and your users’ data), they won’t use it unless you let them.

    Each Google Analytics account has its own privacy settings where users administrators can choose to:
    1) Share with other Google products
    2) Share anonymously with Google and others
    or 3) Not share at all

    That being said, it all comes down to whether we trust Google (or any other company) to follow privacy laws and policies as strictly as they should.

    Another related question I’ve been asked is “can I get my data out of GA if I want to switch to something else?”. Short answer: yes. See here:

    I’ve been using GA for years, and would recommend it for any website.

  10. Denise Eisner says:

    While I think GA has its place (it’s easy to use, custom filters, etc.) and is highly affordable for cash-strapped gov departments, we do have to look at what Google does or doesn’t do to anonymize IPs. I’m still researching this one. In the meantime, take a look at the number of government requests for Google information (sadly lacking detail):

  11. Laura says:

    Thanks folks. All good stuff. Not sure when part 2 will come out but I’ll link it from here in the comments so that you’ll all get an update when I do. Just so you know, the focus of my post will be about when, where and how it can be used, since I don’t think there is an all encompassing yay or nay answer.

  12. DM says:

    Looks like this snippet of javascript will effectively anonymize user IPs before they go to google:


    Linked from:

  13. Laura says:

    Came across an interesting resource today (via Twitter but forget who) a comparison between Google Analytics and WebTrends . The value is in the comments.

  14. Paul says:

    Hi, I think the latest version of WebTrends is hosted only. They no longer have a local version that can be installed. We are switching to GA because of this.

  15. Laura says:

    Hi Paul. This post is really old. A lot has changed since then!

  16. Paul says:

    My oversight Laura, I did not look at the date. Thanks for the clarification. I was actually on your site following links related to the WET project. Great to see the Cdn Gov’t doing something innovative and sharing it with Canadians/World. We will either use WET or jQuery Mobile for our redesign.

  17. dzivkovi says:

    Hi Laura, I also didn’t see how old this post was. It is one of the most interesting threads I was able to Google on the topic.
    So, what is new with respect to GA Privacy issues and difference between GA vs. WebTrends privacy statements?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s