From DM News' Special Report on Web Analytics: Peterson Says Web Analytics Getting Better All the Time
Eric T. Peterson, vice president of strategic services at Visual Sciences, is a veteran of Web analytics and author of three books: "Web Analytics Demystified," "Web Site Measurement Hacks: Tips and Tools to Help Optimize Your Online Business," and most recently, "The Big Book of Key Performance Indicators." Mr. Peterson spoke about the state of Web analytics with DM News Senior Editor Melissa Campanelli.
[DM News' Special Report on Web Analytics is available as a PDF file. To download the article click here or on the image.]/SPAN />
MC Last year, you said that the attention search is drawing is not affecting Web analytics. Do you still believe this to be the case? Are they working together more?
EP I think if anything, a greater number of companies are becoming attuned to the need to measure the effect of traffic coming from search engines, regardless of whether it's paid or organic. By and large, the people I'm talking to are looking well past clicks at this point and focusing on conversion rates for terms and groups of like terms in an attempt to control costs while maximizing value. Especially when companies have an integrated environment, one where you can see your search marketing expenditures juxtaposed against the value said terms are driving, either in terms of direct online sales or against a model designed to estimate the value of a conversion event. I'm starting to see some pretty good decisions being made about how to allocate money and resources to search-related projects.
MC Where is Web analytics today?
EP Getting better all the time in my opinion. The market continues to consolidate, something I predicted would happen last year. The past 12 months saw WebSideStory merge with Visual Sciences, the company I now work for; Unica purchase Sane Solutions; Coremetrics purchase IBM SurfAid; and Microsoft acquire Deepmetrics.
I now predict that the pace of acquisitions will slow in 2006/2007 and this smaller base of companies will further hunker down and work to further define how their organizations and applications will provide great value to their customers.
Also, the "really big" event for Web Analytics professionals has doubled-up: Jim Sterne's Emetrics Summit is being held twice a year now on both the East and West Coasts. The next event will be in Washington, D.C., in October. Go to www.emetrics.org for details.
MC Do you have any updated data on the subject?
EP Well, since I've left JupiterResearch to assume the role of vice president of strategic services at Visual Sciences, objective data is hard to come by. Still, the pace of consolidations I described coupled with the increasing success of top vendors continues to paint a picture of a strong market.
MC What are some of the biggest issues being discussed right now in the Web analytics space?
EP The biggest thing on my radar screen is how companies actually use Web analytics data organizationally. It's one thing to make the investment in technology and another to put that investment to work. The companies that are really making noise about their use of Web analytics data are the companies who are taking a strategic approach, looking to integrate the technology into the organization and not just hoping that some amazing result or insight will just jump out at them.
One response to this issue is the emergence of key performance indicators as a strategy to distill the great volumes of data generated by Web sites and Web marketing efforts into a reasonable set of business metrics that can be widely shared and understood. In fact, my most recent book, "The Big Book of Key Performance Indicators" is a direct response to companies asking "which metrics and reports are most important?"
MC Is consumer cookie rejection and deletion still a worrisome trend? Is it more worrisome or less than last year? Why?
EP Yes, but unfortunately many companies appear to have simply adopted an "it happens" mindset. Given established methodologies for measuring cookie deletion it's interesting that people would still just look the other way.
I haven't seen any new data on cookie deletion rates recently, or at least nothing that adds anything meaningful to the conversation, so I'm not sure if the problem is getting better or worse.
MC In your opinion, is it better to use first-party cookies over third party? What's the difference between the two?
EP First-party, absolutely. A first-party cookie is far less likely to be blacklisted by anti-spyware vendors and thusly recommended for removal.
Both types of cookies work the same way, but given what we know now, in my opinion companies should use first-party cookies to minimize inaccuracies arising from cookie deletion and blocking.
MC Is the cookie at the heart of Web analytics? Will it always be?
EP Yes and yes. Good Web analytics depends heavily on being able to both sessionize visitors and keep track of their behavior from session to session. The cookie is the easiest way to do this, the strategy that works in most situations despite cookie blocking and deletion. I've looked at this problem a lot in the last 18 months -- surely more than is healthy at this point -- and I just don't see a replacement for the browser cookie.
If you have a "better" visitor identifier, like a unique login or authenticated username, use that! But recognize that this "better" UUID is still going to be tracked using a cookie in most instances.
MC Have you come across any advances, analytic tools, or technology worth noting?
EP Well, I left JupiterResearch to go to work for Visual Sciences so I suppose that says something about Visual Sciences technology. I actually work for another former technology analyst there, Bob Chatham from Forrester Research. Beyond that I'm biased and so it's probably better that I don't answer.
MC Whom would you consider as among the leading Web analytic firms?
EP Again, I'm biased in this regard so it's better that I don't answer.
MC Can marketers do Web analysis in-house?
EP Absolutely, and they should. Who better to own the very data that describes the success or failure of marketing efforts than marketers themselves? That said, my standard recommendation is that marketing groups hire experienced Web data analysts and give them enough autonomy to objectively evaluate marketing efforts. This autonomy cuts down on the tendency that some companies have to hunt for data that validates a pre-conceived notion of how a marketing program "should" perform. Web analytics works best when measurement expectations are clearly defined in advance, not after the fact or on an ad hoc basis.
MC The prevailing technology seems so similar. How do vendors differentiate themselves?
EP I still think that quality of support is a key differentiator in the vendor-sphere. Take Google Analytics as an example: all-and-all the Urchin application is good and provides many features for free that other vendors sell. Unfortunately, dedicated support for Google Analytics appears to only be available to Google's largest customers; everyone else either has to cull through online documentation, hire an external agency, or ask questions in the Web Analytics Forum I founded a few years back (groups.yahoo.com/webanalytics)
For some small companies this probably isn't a huge problem. Given enough time and focus you'll probably get the answer your looking for. But what if you need the answer now or don't have the resources to poke around on the Internet? Or what if something goes wrong with your data collection? In situations like this I think companies benefit greatly from having a named account manager they can e-mail or call and get an answer.
MC Which industry categories are using Web analytics to their advantage and which are lagging?
EP Retail and advertising sites still lead the pack in their use of Web analytics data as best I can tell but I think as more companies begin to hire dedicated Web analytics resources this will probably change. I personally believe that any company with sufficient motivation can successfully leverage Web analytics to create a profound impact on their visitor's experience. All that is really required is a strong desire to improve and the willingness to act.
MC Is knowledge of consumer behavior online helping retailers format better campaigns and tailor offers?
EP I think so, but I suppose it depends on your definition of "better". If by "better" you mean campaigns that bring the company more customers, more loyal customers, or more high value customers than I think that the emerging segmentation technology offered by a number of analytics vendors is allowing marketers to generate a more refined picture of their visitors and customers. The best retailers are rolling these insights back into their campaign efforts.
One thing that I've been discussing recently is the idea of using the Internet to soft-launch and test offline and multichannel campaigns. I think that testing response to creative online prior to spending millions on print and traditional media campaigns has tremendous potential. You have a relatively low-cost environment where you can observe visitor behavior and actually ask people what they thought about the campaign. Why wouldn't you want to do that?
MC What best practices would you recommend for marketers?
EP I think the most important thing I've learned over all my years in this space is that in many instances the process a company applies to their use of Web analytics is more important than the technology they deploy. Marketers need to get used to the idea that Web analytics doesn't "just happen" - people make it happen - and that the real value from any investment in Web analytics doesn't come until you start experimenting. I'm a big fan of controlled experimentation and I believe that CE is as fundamental to Web analytics as tagging pages or grepping through log files.
Oh, and don't forget to read my books "Web Analytics Demystified", "Web
Site Measurement Hacks" and "The Big Book of Key Performance Indicators", all available through my Web site, http://www.webanalyticsdemystified.com/
MC How should vendors better serve client' needs?
EP Universally we need to continue to listen to our customers and provide consistent, relevant guidance. Especially as more companies make investments into "Web 2.0" technologies, the vendor community will need to re-examine the guidance we've given in the past and evaluate which of that still stands and which needs to be revised.
MC What's the risk if you don't have a Web analytics program?
EP Same as always, you don't know what you don't know.
MC Why the need for Web analytics?
EP Again, how do you know you're successful if you're not measuring your success? Web analytics and Web measurement technologies when thoughtfully deployed can have a great and profound impact on any online business. If you're satisfied being average, be average. Most companies I work with strive to be great and they see their investment in Web analytics as a competitive edge that will help them achieve greatness.