Rand Schulman and Eric Enge talk about Web Analytics
Published: July 16, 2007
Rand is considered by many a new marketing and media pioneer and innovator. In 2005 he was named one of the Top 100 BtoB marketing executives by BtoB Magazine. He is regularly quoted in the WSJ, Investors Business Daily and Business Week and in numerous marketing & media publications, including AdWeek, and Advertising Age. He is a founder and past board member of the Web Analytics Association.
In his twenty years as a high technology executive he is credited with many transformational products and solutions; from the early computer-based paint and draw systems on personal computers to creating one of the first DTP products in 1986 at Island Graphics Corporation. He founded and was CEO of Keylime Software which is credited as one of the first ASPs; headed product and strategy at WebTrends for several years, and was the CMO of WebSideStory until 2007. He has led world known brands at Software Publishing/Harvard Graphics, and built interactive agencies at Vizicom. Rand is currently the General Manager of Internet Products for Unica Corporation (UNCA).
As a regular speaker, Rand is well known by both technology analysts and financial institutions. Rand has been an executive officer on two IPO teams (TriTeal in 96 and WebSideStory in 2005), and has seen two of his firms acquired (Island Graphics and Keylime Software). He has also spearheaded the acquisition of numerous companies in the last five years. Rand has extensive experience with Venture Capital backed organizations and has worked with many tier one VC Firms including: Battery, Enterprise, Paragon, Sorrento, Summit and TA.
Eric Enge: Can you provide us a brief overview of your background?
Rand Schulman: Sure. I am a marketing guy. I joined Keylime Software in 1997. Keylime was ultimately acquired by Yahoo (YHOO) in 2001. Before that, thirty years ago in San Francisco, I was working for a major online/offline advertising agency. We would have creative sessions such as copy sessions, or sessions around graphics and various components of offline display advertising. We would always have these sessions and everybody had their opinion about what the best color was or what the best call to action was or what the best copy was. A lot of the orientation around the creation of the creative was; oh, I think thatís good or that color just has to be red, or that color just has to be green. Well, as we know tech marketers started reframing the ways to measure creative in one-to-one marketing by measuring response.
I thought it would be really great to start using the Internet media space as a way to gather empirical data about the effectiveness of all my communications programs. So, we started measuring monitoring and ultimately maximizing, the conversion responses to various components of a web page; check out page.
That was in 1997, and then we had the dotcom turn down, and the notions of simply grabbing more eyeballs, and getting more hits fell out of favor. People woke up and started to understand that the online business was really very much the same as the offline business. And, everybody has heard since that we have developed really four primary site types that mirror the offline business world.
Those are: (1) sites that are geared towards selling more products (e-commerce sites); (2) sites that are geared towards capturing leads (lead-generation sites); (3) sites built around the media model of stickiness and getting users to return to the site more often and stay longer so that one could present advertising; and: (4) self service sites, where the principal benefit is to reduce cost. or retain a customer loyalty by making the site easier to use.
Examples of self service sites are airline sites, hotel reservation sites, and banking sites. Another example of a self service site is the internal intranet self service site. On those sites, we are trying to help optimize knowledge workers by simplifying access to knowledge bases.
Eric Enge: Can you talk about how the market shifted in the analytics world?
When I presented to people, at venues such as Internet World, I used to regularly ask how many people were from the IT department, and how many were from the marketing department, or the business owner. From 1997 to 2000, it was hugely IT oriented and I would do five to seven of these conferences a year. Around 2001, it started going from 80-20 in the IT direction, through 50-50, and then clearly to 80-20, in the marketing / business owner direction. We saw far more marketing people embracing analytics. The issue with marketing people embracing analytics is that marketing people come from a core creative discipline, and itís generally true that creative people are not driven by numbers.
Now we have started to see a hybrid marketing person emerge, where they were are creative, but also have the ability to understand numbers. These people are very much like engineers. They understand that they have to use the tools at their disposal. We are going through this metamorphosis, where those two individuals are becoming one individual. I think there is a greater understanding by the marketing world out there that there is this merging of individuals, and that schools will start embracing the notion of web analytics and digital marketing. This will further the science of marketing as it continues to cross from the offline channel to the online channel, and then back again from the online channel to the offline channel.
Eric Enge: Has this started to become a factor in education?
Rand Schulman: Sadly to say it hasnít. I have met with many deans of many business schools, and the deans are often driven by seeking to endow chairs or build wings from or get large grants. They are driven by those donors who havenít considered the tools that exist out there. Unfortunately, I expect that the business schools will only start embracing this when people that have made their money by building companies that sell digital marketing tools start donating money to the business schools. Then they will carry the curricula, but this may be a decade or more away.
Eric Enge: Letís talk a bit about your time at Web Trends, WebSideStory, and your involvement with the WAA.
Rand Schulman: I left Keylime Software in 2000, and I took sometime off. I got a call from WebTrends, and NetIQ had just acquired WebTrends at that time. WebTrends had just gone public with a very large market cap, and NetIQ bought them for about a billion dollars. I knew at the time that ultimately WebTrends should be sold to marketing users and business users, and I knew that NetIQ really didnít have any reason to own WebTrends. It sold to (IT) buyers of IT solutions and was certainly a great company. But, the two did not fit and there I was, the Senior Manager of the business responsible for their products at the time. Those products at the time were built on six separate pieces of architecture starting with log analyzer and at the time we had a midrange product called WebTrends Reporting Center, and we had a high end database product that we called WebTrends Intelligence Suite.
We also had the same thing in on demand offerings. But we had a large number of developers that were working on six separate architectures and six separate products. We created ďProject HoodĒ to align those products under a single analytics engine, and I think they have done a really good job of pulling that off. We had to reconcile code bases and pricing paradigms, and that was challenging.
I went to WebTrends to spin it out, and I was very vocal with the NetIQ management at that time. The two should never have been together, and I believed that they could unleash the value of the division by spinning it out, but they couldnít do that during my tenure at WebTrends. They did complete a spin out right after I left, once they brought in Greg Drew as the CEO.
At the same time WebSideStory had made me an offer to come to become CMO . They had just hired a new CEO at the time, Jeff Lunsford, and he and I started pretty much on the same day and that was in March of 2003. Right about that time all the vendors noted that online business is very similar to the offline business, and we started talking about four types of businesses similar to the list of web site types we already discussed. All of the major vendors started using online and offline business analogies to communicate to web analytics and digital marketing professionals. WebSideStory executed pretty well around those concepts and the business took off. We completed a public offering about a year after I joined the company, and thus we had some equity and a way to acquire other firms in the digital marketing space. One of my major contentions was in 2003 that web analytics is not inherently a stand alone application. You canít just look at the numbers, but you must take action on those numbers.
WebSideStory started building and acquiring other online applications which would take action from the data that was collected and analyzed in the web analytics applications. We acquired Atomz, which had a very good reputation with a site search product and a web based content management product. We merged those products with the web analytics products to allow the analytics data, based on business rules, to start to take action.
We also built a suite of APIs that could be integrated with other applications and their data, We called it the Stream API Platform. We made it available to end users to integrate their data, and to other vendors in the ecosystem, such as the ad serving guys, the email guys, and the behavioral targeting guys.
I think at the time WebSideStory was the first mover around the integration of actionable applications, and the creator of the first suite of standardized and public APIs
Eric Enge: What happened when WebSideStory acquired Visual Sciences (VSCN)?
WebSideStoryís acquiring Visual Sciences was quite controversial within WebSideStory. It was not completely embraced, because Visual Sciences is, more or less, a closed business intelligence system that does some things very, very well. It is true that you can get data out of the system, but itís very costly to do that. Itís an elegant application for slicing and dicing huge amounts of multidimensional ďrealtimeĒ data, but marketing people just donít need the infiniteness of the slicing and dicing of real time data. What they need is ability to see enough to take actions on that data, and get some feedback from the actions of the data, before they continue to any further granular slicing and dicing of that data.
The acquisition of Visual Sciences was not completely embraced by the entire management team, and employee base of WebSideStory. Consequently you saw a departure of, many Senior Managers and employees from Visual Sciences, and it seems they have refocused the company around the new mission, targeted more towards the generalized horizontal analytics user .
Eric Enge: Itís also interesting now that they have re-branded to Visual Sciences as a company.
Rand Schulman: Basically, it says that everything is visual. If you agree with my primary contention that web analytics products collect data, and then they trigger other applications, there is no ďvisualĒ requirement about a triggering mechanism whereby you threshold the business rule, and you take the actions. Itís a very powerful application, and very complimentary to some business intelligence solutions. But, itís very, very different than the applications in the markets that we identified early on.
Eric Enge: One of the big things in analytics is getting an ROI from your analytics investment. You are not getting ROI by looking at visitor numbers and page views, you have to dig in and do some real analysis, and you have to tie it to the business goals of the company. In fact I was talking with Eric Peterson last week, and he put it very well, His notion was that if you arenít doing situational performance testing, like AB testing and multivariate testing, then you arenít doing web analytics.
Rand Schulman: I completely agree with you.
Eric Enge: My guess is that a very large percentage of people who use analytics packages donít do those things currently.
Rand Schulman: I agree with that. I think there are very few marketers that are engaged in what the scientific method. Engineers and scientists are trained on the use of scientific methods, but marketers are not. The scientific method will tell you youíve got to have that control group. You have to understand what a control group is, what multivariate testing is, and what AB testing is and why they are fundamential to success in marketing.
This leads me to the point of why I joined Unica. The market is going full circle to encompassing cross-channel marketing and cross-channel action, not just cross-channel reporting, but cross-channel action, and combining online and offline action. And, thatís why I came to Unica, who I believe is uniquely suited to that.
We need to decide what is the criteria that describes cross-channel marketing . How does the cross-channel differ from multi-channel, and how can you start taking a cross-channel action, not just cross-channel reporting on the online side, or not just cross-channel reporting on the offline side?
Eric Enge: Did you have any other thoughts on the WebSideStory re-branding to Visual Sciences? What is it that causes a company like that to leave behind a well known and respected brand, and switch to another one? You touched on part of it; with the suggestion is everything is Visual, but the other thing that happened is that they lost the word Web.
Rand Schulman: Yes. They did lose the word Web, and they lost the word Web for two reasons. It is not just the online channel, but secondly it doesnít have a thing to do necessarily with marketing at all. Itís a business intelligence tool, and it may not be looking at channels at all. Itís a data segmentation engine; it collects data, and it slices and dices that data, and it presents for you the results of that analysis. Visual Sciences was funded by In-Q-Tel, which is the VC organization for the CIA. They are well suited for looking for those needles in haystacks in real time, such as key intelligence information in email traffic and over phone lines.
Thatís truly Visual Intelligence, where itís probably best represented visually and they do a compelling job of graphically representing lots of complex data, but for the most part, itís over kill for everyone but the highest level of marketing analyst . So, they have re-focused the company, and itís too bad because they have ceded the market to Omniture (OMTR), WebTrends, and Coremetrics for the high end e-tail space. At Unica, I head the Internet marketing solutions group, and our goal is to produce the best web analytics product out there, but also have a suite and ecosystem of integrated digital marketing applications, and, to make sure that those are in tune with our offline applications that Unica knows so well. That is our differentiator, the fact that we are truly integrated online and offline and that we can take actions with our applications: manage leads, campaigns, and marketing plans, in both worlds.
We are a kind of a game changing company, where Omniture and WebTrends are focusing online. .
Eric Enge: You came at it from a different direction, didnít you?
Rand Schulman: Yes. We came at it from the offline side. But, all the vendors are talking about this ďcross-channelĒ thing now, and what they are really talking about is cross-channel reporting. They are not talking about cross-channel actions, and thatís I think the next five years is going to see more and more companies follow Unica, and get into the cross-channel actions both, on the online side and the offline side. Again, WebSideStory was the pioneer, online company allowing applications take actions on the web analytic data on the online side alone. Now, Unica is enabling companies to do that across both channels.
Eric Enge: What are your thoughts on Eric Peterson going independent?
Rand Schulman: Eric Peterson and I have been in two companies together. WebTrends, and WebSideStory, and we certainly worked together very closely when he was an analyst at Jupiter. We know each other pretty well. Eric and I donít always agree on everything. I come from a pure solutions point of view, because my DNA is in marketing. I know what I want to achieve with these applicaitons, and Eric has an approach to these things more from a technology point of view. But, more and more, his point of view and mine are becoming very similar. I respect him, because he will tell you what he thinks. Weíve had our differences, but he and I are good friends, and I think itís good to have differences. It makes the world go round. I think he is going to do really well as an independent by the way.
Eric Enge: What about Avinash Kaushik leaving Intuit, and becoming the Google Analytics evangelist.
Rand Schulman: I just talked to Avinash two weeks ago at Emetrics. I had lunch with him. He is a great presenter, he is hilarious, and he is a great guy. I think we need more guys like Avinash,Bob Page, John Marshall, Bret Hurt, Bryan Eisenberg, Jim Sterne, Eric Peterson, and Josh James really evangelizing what we can actually do using these kinds of applications. I think Avinash is another great example of that kind of person.
Eric Enge: Can you comment on the accuracy of analytics solutions?
Rand Schulman: We have to see the forest for the trees, we have to apply the eighty-twenty principal, and understand that analytics provides great trending information that will empower other applications to optimize other marketing programs.
Eric Enge: Right. So, when you refer to trending information, what you are really trying to say is we know that we donít have perfect measurement. We just know that if we have a measurement, and then we see changes from one week to the next, we have meaningful data that we can really deal with.
Rand Schulman: Thatís exactly right. I have often instructed our sales teams not to try to sell the customers who own other web analytics products. And, even when those customers come to me and ask whether or not, they should buy the new products, I often tell them no, they shouldnít, because it will introduce a whole new set of variables to try to optimize. Itís often times at odds with the goals that they are actually trying to achieve with the use of these products. So, I usually counsel them to stick with their current vendor, and just to try to get better service, or a clearer perspective, from the products that they are using already. I know thatís a heretical thought,
Eric Enge: I was also interested to see Google Analytics come out with a new release. They rectified a couple of gaping wounds, with one of those being the lack of ability to email out reports, and the other was not having a dashboard system. Do you see the continuing progress of the lower cost vendors as putting pressure on the higher end vendors?
Rand Schulman: Not really, just a little bit. As you said Brett and the Google guys are filling fundamental holes with reporting and the lack of dashboards that they were missing before. Their historical users, the Urchin users, had not required it. Again, I think Google (GOOG) has to apply the 80-20 rule. I think it would be counterproductive for Google to give away more advanced functionality, because, frankly, they can't support it. You need some advanced users to utilize it and to know what questions to ask. You see, all analytics products have all the answers, -- we just donít know what questions to ask!
But, I think Google could probably do a better job of integrating the analytics applications with the other applications that they have, and measuring and maximizing other applications that they have in their stable. So, there are lots of things they can do with Google Analytics to apply it horizontally against all the other Google products. But, I donít think that they want to keep developing it against a deeper and deeper set of functionality to go after the low end marketing analyst in the long tail.
Eric Enge: What about the challenges with tracking traffic and being able to decide on actions for mobile devices.
Rand Schulman: Thatís a very good question. In fact let me expand the question to ďWeb 2.0 applications.Ē There is some similarity in these, and one of the pieces of similarity is that we are going to see more monitoring going back to the server side, not so much on the client side anymore, because the client side is going to be ultimately little mobile devices.
Rand Schulman: Probably not. So, you are going to have to measure things that are coming in and out of the server. Thatís going to be a fundamental shift back away from the client side to the server side, where log files and tag information will come together. To optimize this mobile application, we are going to have to monitor a streaming application, and to do this we are going to require both client side and server side monitoring . Secondly, we are going to have to have a notion of the correlation of orthogonal web-based eventsĒ versus the linear display of a series of page view- based events.
Over the last decade we have developed features and practices around the optimization of hierarchies of pages. When you use a model which is not based on presenting page-product campaign hierarchies, it becomes more challenging to optimize a web based event, where windows are popping up all over the place, and I think we are going to have to be more mindful of optimization of these web based events, and we have to be very careful about both drawing a correlation between these web based events and page views, and a correlation between web based events and engagement metrics, and conversions where greater engagement does not necessarily equal greater conversion. We just have to be very careful about those kinds of judgments and analyses.
Eric Enge: Right. We also have to be mindful that the methods for monetizing different environments maybe completely different. Itís monetization that people want to track, regardless of whether itís a cost savings or a revenue event or a lead. They want to understand how they are affecting it when they do things.
Rand Schulman: Thatís right. But, I can show you some studies on this, where greater engagement actually equals lesser conversion. It depends on the presentation of the campaign, or the device that the user is interacting with. It depends on the company and what you are asking, but a lot of people assume these days that greater engagement equals greater conversion, and thatís just isnít the case.
Eric Enge: Yes. Thatís the reason why you need to have a method to analyze and test and analyze and test etc.
Rand Schulman: Thatís right. Ultimately your goal maybe lesser engagement, so you just have to know what.
Eric Enge: I will give you an example. Letís say you have somebody that knows they need a part to repair something, maybe itís a car. They just want to find the thing and buy it. You donít want to engage them in an article about how to install it, if they actually at that moment want to make the purchase. Itís a more efficient use of their time to make the purchase, get the part and they are done, it takes them two minutes. Thatís the experience they want.
Rand Schulman: Yes. There are also some manufacturers out there that are asking customers on the website to configure their product. Some of these are finding out that the clientís greater engagement on the website means lower conversion to their store. The user is getting their curiosity satisfied on the website. As a result they are not making themselves available to be sold to offline by that sales person who has a very high conversion rate compared to conversion rates of web results. So, we have to be very mindful of using our new tricks.
Eric Enge: Right. Are there any other major challenges facing the analytics industry other than those of Rich Media and Mobile?
Rand Schulman: I think yeah. The general issue of privacy will continue to raise its ugly head. My feeling about privacy is that bad actors will do bad things. We have laws to prevent bad actors from doing those things. Scott McNealy said it best regarding this issue, ďget over itĒ, but I think itís going to be a real challenge for most Americans to get over the issue of privacy. When you start seeing the combination of companies like Google and Double-Click, you start understanding the patch work and patterns that can be brought out by stitching together the databases that are contained in Google and Yahoo.
I think that Americans and/or Europeans, and most freedom loving societies, have a real aversion to the invasion of privacy. Yet we have these huge and vast stores of personal information that are going to be employed for the benefit of us all as individuals. Those benefits will help us live richer lives. So, I think thatís probably the biggest challenge, we have to face.
Eric Enge: Great, thanks!
Rand Schulman: Thank you.
- Steve O'Brien - Apr. 23, 2007
- Bryan Eisenberg - Mar. 24, 2007
- WAA's Brian Induni - Mar. 11, 2007
- Clicktracks' John Marshall - Feb. 28, 2007
- Avinash Kaushik - Feb. 25, 2007
- Eric Peterson - Feb. 11, 2007
- Jim Sterne - Jan. 17, 2007
- IndexTool's Dennis Mortensen - Jan. 8, 2007
- Google's Brett Crosby - Jan. 3, 2007
- 10 Cool Things You Can Do With IndexTools
- 2007 Web Analytics Shootout - Interim Report
- Web Analytics and Cookies
- Brett Crosby, Avinash Kaushik - June 13, 2007
- Brian Induni - May 15, 2007
- Dennis Mortensen - Mar 9, 2007
- John Marshall - Mar 8, 2007
- Jim Sterne - Feb 22, 2007
- Avinash Kaushik - Feb 9, 2007
About the Author
Eric Enge is the Founder and President of Stone Temple Consulting (STC). STC offers Internet marketing optimization services, including SEO, Social Media and PPC optimization, and its web site can be found at: http://www.stonetemple.com.