Adam Jenkins is an analytics consultant working in NY. He is a husband and father of 2 little boys. He loves baseball, sailing and beat up old Mercedes diesels. He has worked on analytics data sets in Travel, E-Commerce, Publishing, and Manufacturing. If you cut him, he bleeds Omniture Green.
1. What got you started in web analytics?
AJ: I have always loved to take things apart and figure out how they work. When I was a kid, this trait got me in loads of trouble as I dismantled clocks, watches, computers, radios and other things around the house, which usually never worked again! After college I started working as an analyst and kept asking why, to many people’s annoyance. Eventually someone designated me for a job looking over an Omniture Implementation to see if the conclusions of the analysts fell in line with how the data was being fetched on an e-commerce site. I immediately fell in love with the cause and effect nature of web analytics and the fact that the job was never the same two days in a row. I have been working in the field ever since.
2. As a consultant, what do you enjoy most about your job?
AJ: I enjoy the fact that I have been able to work with clients in various industries and the unique challenges each has brought with them. For instance, when working in publishing, the goals and measures were vastly different than when working with an E-Commerce client and so on. I really enjoy the fact that my job is never static and 90% of the time, we are working on an issue that is new to everyone involved. I cannot understand how people go to work each day and do the same thing for 30 years; I would go mad very quickly.
3. I’ve been seeing a lot of companies doing “re-implementations”, how can companies best update their implementations without starting over?
AJ: This is a tough question. Many companies are finding themselves overloaded with data after a decade or more of focusing on “getting data” rather than focusing on understanding the story of the data. Unfortunately, many of the companies are measuring many things that may not be remotely associated with what they need to define success or failure. Years of development “band-aids” being applied over existing “band-aids” have left a pretty scary looking solution that in many cases is no longer scalable or relevant to the current course of the business.
I find that asking questions is always a good place to start, and three questions seem to open the floodgates for feedback.
#1. What is the most important business question you (Client) are asked to answer by your organization (Boss)? (Month over Month, Quarter over Quarter and Year over Year)
#2. What reports are currently being used for the analysis behind answering question #1?
#3. What question are you most commonly asked that you cannot answer?
Usually the feedback to those questions or ones like them, are a good guide to find out where the implementation is “fat and happy” and where it looks “emaciated”. Most of the time, this exposes custom variables that may be re-purposed, naming conventions that are outdated and not scalable, and campaign tagging processes which have not evolved with the business.
4. When is it time to blow up an implementation and start over?
AJ: I usually start thinking Ka-Boom when the proposed solutions require workarounds. I have to admit, I usually start thinking Ka-Boom very early in most situations where very little documentation exists, or the churn rate has been very high. In some cases, an implementation can be fixed but in many it cannot.
5. If companies do decide to do a clean re-implementation, what are the typical issues they face?
AJ: Lots of times, historical data becomes very difficult to map from “what was” to “what now is”. This can be tough, but I try to remind people that the sooner the new solution is implemented the sooner they will have period over period comparisons available. YoY is a great measure, especially when you work with a business that swings wildly based on seasonality. Many folks are reluctant to lose that insight, even if it is horribly flawed. This is where logic and reason are your arguments best friends.
6. With Social Media, it’s easy for anyone to be a “guru”, how do we as a community better police “gurus” who do poor work which undermines our industry?
AJ: I have no idea but I am concerned that a few bad apples can spoil the bunch. I guess companies can be encouraged to require objective certifications and such. Perhaps we can set up a peer review site? I really don’t know. I myself have run into issues, in the past, of being cocky, and thinking in a headstrong manner and it is not the best path for my work or my clients. Most importantly it is not the best path as an ambassador to our field. It seems like the true “gurus” in this field, at least the ones I try to emulate, (Ben Gaines, Michael Halbrook, Adam Greco, Derek Tangren, Rudi Shumpert, Jeremy King and the gang of Analytics NINJAS that hang on the twitter playground) all seem to be humble and have more questions than answers.
7. I’m a huge fan of Omniture Discover but I’ve never had the opportunity to play around with Insight, what is it about Insight that you like so much?
AJ: This question made me immediately think of Almost Famous
William Miller: So Russell… what do you love about music?
Russell Hammond: To begin with, everything.
EVERYTHING. Insight is a powerful tool which humbles me daily. Marketers love the visualizations, but I love the ability to instantly filter the entire data set based on dynamic selections. You can fetch powerful segments to target market, you can create customer personas, you can spot software glitches, you can relate your marketing to your bottom line, you can push all your call center and customer service data in one place, you can do almost anything with this tool. You can put almost any type of data into this beast and relate it to whatever you want. Basically every data set can be anchored to whatever field the user wants it to, with some tweaking from your friendly Adobe Consultant. You can shift the levels of the data. In other tools you are limited to Page View, Visit or Visitor level data but in insight you can add additional levels. I have seen Web Data be paired to Point of Sale Data to provide targetable segments for various marketing channels. I have seen companies spot internal software glitches that no user group had previously exposed. I cannot say enough about this tool. The coolest thing to me is that I have a limitless sandbox for discovery with this tool. This tool makes me proud to be a geek.
8. Do you have any specific examples of how you have used Omniture Insight that has proven to be a big win for a client?
AJ: I am not sure I can answer this on behalf of clients. But I do think some pretty cool case studies have been submitted to Adobe.
9. In your experience as a consultant, have you noticed that specific departments within an organization tend to be better owners of analytics than others?
AJ: This is a great question. I have seen the most success when Marketing owns analytics. But my background has typically been more on the business side of the fence. I am currently in a more technical role, so my opinion may shift!
10. What advice do you have for someone who is interested into getting into web analytics consulting?
AJ: Ask as many questions as possible so that you can explain your problem and solution to a five year old. Take comfort in the fact that the data will argue for you!