A short while ago our own DavidF scored an interview with Eric Peterson of webanalyticsdemystified.com. For those who dont know him, Eric Peterson is an analytics consultant, keynote speaker, and author of numerous books on web analytics, such as: Web Site Measurement Hacks, The Big Book of KPIs, and of course Web Analytics Demystified. When not writing he acts as a consultant for the likes of PBS, the New York Times, Cnet, and The Wall Street journal. He is also a major proponent of free information, giving all of his research, his presentations, and his white papers away for free on his website.

What do you think the biggest change has been in the Analytics Industry in the past twelve to twenty four months?

I think that the biggest change we have seen in the industry is that there has been a dramatic surge in the interest in the study of web analytics primarily driven by the free Google Analytics service. Hundreds and thousands of companies around the world are looking now at web analytics that would not have before simply because there was a cost barrier in place.

What do you think is the biggest change that will be coming in the next twelve to twenty-four months?

The biggest stories right now are the newly improved Google Analytics version 3.0 product and the economy.  In October Google announced a major upgrade to Google Analytics and it now includes pretty good segmentation features, custom reporting, some cool visualization options, and API-based access for a handful of companies participating in a private beta.  The net of GA 3.0 is that fewer companies need to look at for-fee solutions to be able to really get into web analytics so it will be interesting to see how that all plays out in 2009. Your readers can get more of my thoughts at my blog:

Also, the economy is starting to create a drag on the industry as a whole, costing some web analytics professionals who have a ton of experience their jobs.  This is pretty unexpected since up until pretty recently companies were clamoring to find this same professional and paying big bucks for the talent in many cases.  I wrote a post about web analytics and the recession economy that gathered a ton of feedback that your readers may be interested in at:

Quality Assurance of one's analytics implementation and the integrity of the data that is collected still seem to be a struggle for many companies. Have you seen companies successfully address this issue and instill confidence within senior management that decisions are being made on quality data?

Yes I have, in cases where companies have focused on the right technology, the right people and the right process. In the absence of good technology, there are always questions about the data. In the absence of people, there is nobody to ask the question to and in the absence of process there is no process to determine whether or not the data should be trusted or not. 

Quality assurance and data validation is probably the single most important reason that I have been railing on companies to focus on the process of doing web analytics for years now and still some companies haven't listened. They still have data integrity issues, they still have questions and they still say "Can we trust this data"?

Your readers can see a whole presentation about how web analytics is done right in the video made at IMC 2008 in Vancouver:

I have heard recent talk in the analytics field about a fear that there will be a back lash against the use of cookies, and that this may bring an end to their use at some point.  What are your thoughts on this, and how would the industry evolve and adapt should this happen?

Cookies are an interesting question. I first wrote about cookies in 2005 while I was at Jupiter Research addressing the decline in the persistence of cookies. Specifically with the launch of the platform for Privacy preferences, and a change in Microsoft's approach towards the Platform for Privacy Preferences (P3P) and all of a sudden cookies become less good indicators of the longevity of a visitor.

Now we have Microsoft IE 8 coming out with "in private browsing" which is a single button that sessionizes all cookies. You won't be able to identify those returning visitors which will impact lifetime value calculations and most other visitor metrics. Apparently, a similar piece of functionality has been in Safari for a while and is now going to be in Google Chrome as well. Whether or not a consumer will use these features is hard to say.

The problem, as my friend Bob Page has talked about, is a Data Chernobyl--the ultimate privacy incursion where some hacker steals Google's data or Omnitures or WebTrends data and makes that information publicly available. If people think and people are aware that they can be tracked on the internet, it may scare them to use more of those features and more of that functionality and thus cookies may decline.

The problem with declining cookies is that the alternatives are likely to be seen as equally invasive and suffer from the same kind of consumer backlash and consumer paranoia. Fortunately it hasn't happened yet and there hasn't been a massive meltdown.  But again, it's one of those things that some people in the industry always walk around waiting for, wondering when it's going to happen and not so much if it's going to.

Have you done some research in the past on salaries for professionals in the web analytics field? How do you think salaries and job openings in the web analytics field will be impacted by the current economic slowdown?

It's funny.  When you first asked me this question we had not seen salaries or opportunities impacted very much by the slowdown. The need for dedicated staff for experienced resources to run these technologies is so pronounced, so important that taking web analytics staff off the table is a horrible idea. You're still paying for the technology, you're just not getting as good of a use for it.

Unfortunately, in the last few months this has all changed.  An increasing number of open requisitions for web analytics experience have been frozen or taken off the table entirely and I hear from more and more experienced practitioners every day who have been laid off (or feel like they're in danger of having this happen.)  It's a shame since especially when companies have found (or trained) folks to do web analytics work, letting them go will likely set efforts back a great deal.  This is a pretty short-sighted mindset in my opinion, but I'm also too close to the sector to be truly objective.

Sadly without much effort I predict that the job situation is going to get worse before it gets better.

What advice do you have for firms that are currently looking for an analytics vendor?

If you are brand new to web analytics it's really simple: get Google Analytics. It's totally free, it's totally great and they have a wonderful team of people supporting it. There's tons of free literature about it and there's nothing to stop anyone from going and getting Google Analytics.

If you are not new, then you can look at the Omnitures, WebTrends, Coremetrics, and Nedstats of the world. All of the paid vendors have a good refined understanding of what they need to do to support their customers, and by and large I think they do a good job of that.

It's a great time to be getting into web analytics. If you're just getting started there are great free solutions, which are a great starting point. Good data comes out of that. It's not like years ago where you had to commit tens, if not hundreds of thousands, of dollars.

Would your advice be different for clients adopting analytics for the first time verses looking to switch vendors?

Certainly if you are looking to switch vendors, there is a reason you're switching: You're looking for additional functionality or maybe better support than you are currently. Maybe you adopted analytics years ago and you're paying higher than market price and now you are looking for a lower price. There are always drivers behind that switch.

The important thing about switching is to look at yourself, to be introspective before you switch. Years ago, everyone said that we are not being successful with web analytics let's blame the vendor. "You know it must be vendor A's fault, let's get vendor B." Well they got vendor B and ran vendor B for a couple of years and they weren't successful with web analytics. The said "It must be vendor B, let's get vendor C".

It turns out it really wasn't the vendors at all, it was the companies and their lack of understanding in what is involved to be successful with web analytics.  I say this all the time, but people, process, and technology are all required, not just technology.  Now, in 2008, there are hundreds of case studies where companies are increasingly successful because they've finally put this together.

What tips or resources can you provide to those who are currently starting to look at the use of key performance indicators?

I would be remiss if I didn't shamelessly plug my book, The Big Book of Key Performance Indicators available for $19.99 US at webanalyticsdemystified.com. It's a good starting point. I outline a hundred odd key performance indicators, talk about the processes of creating KPI dashboards, and offer tips to make these projects successful.

There's a Yahoo group I founded in 2004 for web analytics where the conversation comes up from time to time. There are lots of bloggers out there like Gary Angel, Judah Phillips, and my friend June Dershewitz who talk about the subject from time to time. Also the vendors themselves have focused on simplifying how the data is presented to management and to the broader organization through their support of dashboard functionality.

The good news is that there are lots of places to turn. I wrote the KPI book years ago because there was really nothing to look at and now that that gap is filled in, not to say that you not check out the book.

Can you talk a little about how you see analytics evolving as it is used for measuring social media and mobile interaction?

Mobile is the easier of the two to address. There are a number of companies who have sprung up specifically to measure "mobile analytics". They provide some necessary short-term functionality to help companies grapple with some of the complexity associated with mobile phones, carriers, and other technology-related issues.

But I think that ultimately the traditional web analytics vendors will take an even greater interest in this, not just market around it, but really do something significant with it and integrate that into what they measure. So web analytics is not just about websites on notebook or desktop computers. It's about that interaction that you have with someone, be it on a mobile device or a computer or maybe even on a TV. I think that the web analytics vendors will eventually get that. Nedstat from Holland are actually ahead of the curb in their ability to measure mobile devices today and are doing good work with that.

The social media stuff is harder because it doesn't necessarily lend itself cleanly to integration with web analytics. I am a big fan of Twitter. You can twitter me @erictpeterson on Twitter, but I haven't figured out how to efficiently track my Twitters and, more importantly, integrate them into my primary web analytics deployments. I think that that's the challenge.

The notion of the unique visitor gets lost very quickly when you go into this fragmented social media environment. I identify myself from my Facebook account, my Twitter account, but neither of those knows me when I go to their website. There's this disconnection or lack of continuity and that's the problem that really needs to be solved. That's why we see the social media measurement companies, Buzzmetrics acquired by A.C Nielsen as a classic example, whole companies, whole analysis frameworks that are independent of web analytics springing up and actually having some stick in the marketplace.

What would you say are the three most common mistakes that companies are making when it comes to their analytics usage?

The first mistake is too great of a dependence on technology believing that web analytics is easy and the technology is somehow going to magically provide insights and solve problems. 

The second is not dedicating staff, in small companies not dedicating at least half of somebody's time to focus on analytics and looking for those insights. In larger companies, one or two persons should be able to dedicate all their time to looking for those insights.

The third mistake is a lack of adherence to process, a lack of interest in process. Megan Burns from Forester said that for some people process is a four-letter word. The problem is that web analytics is very process intensive but it's not something that we study in college. It's not something that business owners and business managers necessarily get right off the bat.

So you have to focus on people, process and technology. It's a nice platform to talk about because when you talk about the need for people process and technology it reminds companies of every other enterprise software implementation they've ever done. There's some comfort in that, it's just they actually have to do the work as well.

How long before we can automatically pull analytical data into one platform from Social Media and how do you think this will change the industry?

It's a good question to have last, because it's one that I don't necessarily know the answer to. You think about the different technologies that exist to tie us together when we traverse the Internet. It's not impossible to imagine something like Open ID that says that when I go to webanalyticsdemystified.com, I am the same person who Twitters about web analytics and who has a Facebook profile where maybe I'm part of a Web Analytics Wednesday group and take that ID and tie that data together.

The problem is that consumers have this notion of anonymity and this expectation of privacy on the internet. These are two conflicting needs, the need to identify us, the need to remain anonymous. You see consumer desire to remain anonymous in the NebuAd and Phorm controversy where Phorm was using deep packet inspection to see people as they moved around on these networks over in the U.K. There was a huge backlash against that. The flip side of that is what will we get, what would I get if I identified myself at a website and said that I Twitter at this address and I have this Facebook where I Flickr these photos. What would I get and where is the value proposition?

The value proposition to consumers is almost wholly lacking in web analytics. It's unfortunate, hard to turn back the clock of time. Walt Mossberg from the Wall Street Journal said that if you want to set these tracking cookies on my computer, maybe you should pay me. At the time I thought that's a crazy thing for Mossberg to say, but at the same time he's right, consumers need to get value from giving up information. Otherwise what we're left with is an increasing paranoia about how much companies know about us.