Michael Beasley is a User Experience Architect at ITHAKA and Author of Practical Web Analytics for User Experience, which teaches its readers how to use web analytics for questions UX professionals face. Whether you’re measuring a website’s effectiveness or discerning how users are using your website, Michael’s book has it covered. We had the chance to sit down with Michael for a talk about his career in UX, and the future of the industry.
ML: Tell me about a day in the life of a user experience architect.
MB: As a user experience architect, I work with a huge library of academic journals that people inside and outside of the educational system access. With this in mind, the keyword of my day to day life is collaboration. This can take a variety of forms, whether it be facilitating design activities or bringing together the product team and the product manager to discern how to solve a certain business problem. Also, very importantly, it’s discussing how we will measure whether we are actually successful in solving that problem.
The other half of my job concerns facilitating, planning, and conducting research activities. That research may be very traditional UX in the sense that it’s bringing people into a lab or going out and watching them. Often, this is more quantitative work. It can range from testing the effectiveness of a design to looking at what kind of metrics we will gather about a design once it’s deployed, in order to see whether it’s working or not.
Overall, it’s about working with people in front of a white board with a marker. It’s about collaboration. It’s bringing people’s ideas together. That’s a day in my life.
ML: You’ve been doing this for quite some time. What is your motivator for continuing down the path? What keeps you excited about this job?
MB: On one level, I view designing improvements in user experience as important work. The idea of a product’s services or of a website’s software being designed around human needs and capabilities is an important goal. It makes the world a better place to live in. On a more practical level, I really enjoy solving problems. It’s about more than making a wireframe—rather it’s about discerning our problems and figuring out how to solve them. Design is all problem solving from my perspective, and that never stops being interesting because there are always new problems to solve.
ML: I want to ask you about your previous job at Pure Visibility, where you created and led a measurement team. Can you tell us more about what that entails? Do you have any advice for organizations or executives that are currently in the process of building their own digital analytics teams?
MB: Pure Visibility was an interesting position because I started off as a UX analyst. I wasn’t focused so much on design as I was on answering client questions about user behavior, which meant that a lot of my work was done with analytics as the key tool. We were also selling our clients on more expensive projects, such as usability testing or even going out into the field to look at people. After a few years of bringing an analytics tool set into my UX toolkit, it started to be clear for our agency that answering questions about usage should be thought of as one entity. That’s how the measurement team was born. Sometimes we would answer questions about the effectiveness of the marketing activity, and other times we answered questions about what people were doing once they came in from the ad on a search engine. Really, it was all part of that larger, broader activity of telling our clients what people were doing on their website.
There are specialists that work with analytics, but also people who focus their entire career on design or research activity. In the future, I definitely see collaboration between disciplines. I think the future will have a lot more data-informed design, as opposed to design happening in isolation from web analytics data. The question of how to organize that is still ongoing and it’s hard to say how it will roll out.
ML: While working in the field of usability and UX, what has surprised you the most? Have you ever been in disbelief of what data might be telling you because of your gut feeling?
MB: The thing that always surprises me is that nobody is using websites as much as we think they are at a page level. When you look at the average user time on a page, it is always way shorter than you think it is. You spend a lot of time deciding the copy and what words should be used, so it’s hard to imagine that some people are landing on that page, skimming it, and then moving on. I don’t think it really hits people until they see the average user time on a page and realize, “Wow! Most people are probably not even looking at this page in any great detail.” I think the data emphasizes that your site is just one of many out there that people may be using, so it’s important to keep that in mind.
ML: As we look at usability, data, and technology, do you foresee any major changes in the world of analytics to make your life and the lives of other UX people easier?
MB: I’m excited about the integration of web analytics data with other data sources. You have these piles of data— web analytics telling you what people do on the website, or fulfillment tools telling you what people have purchased. If you are a brick and mortar store, you might have tools that describe what people do in your store. Once you start putting all the data together, it unlocks perfect possibilities for analysis.
If you take, for example, integrating CRM data with web analytics data, you can start to identify a particular kind of user that’s valuable, and segment out their web analytics data. Then you can look at what they’re actually doing on the site.
Also, figuring out how to better measure people moving from device to device excites me as well. This is just a thing that people do. If you can’t get people to log into your site, it’s remarkably hard to measure the behavior of a single user across devices. I don’t know exactly how that’s going to get better, but I would say that it’s an ongoing problem that we face from a user experience perspective.
ML: Let’s talk about your book, Practical Web Analytics for User Experience. What motivated you to write this book?
MB: Primarily, there was not a lot of reference material out there for UX people who wanted to pick up web analytics. At a very practical level, I’m trying to help people. Web analytics is a very hard thing to learn. I wanted to provide a starting point for people in the UX world. Unless you came into the field within the last few years, you probably didn’t learn a lot about web analytics. By chance, I got to work at Pure Visibility and was able to pick up web analytics.
Also, user experience, very generally speaking, has different concerns and cares about different things with regard to web analytics. We care a lot about how people are moving from page to page, what users’ behavior at the actual page level look like, and really digging into search queries that people make on the website, for example. Those are the kinds of things we want to dig into.
ML: In the past, there has been conflict over creativity versus analytics. How can people in UX and design space start using data or analytics more, and bridge the gap between those two worlds?
MB: First, if you’re the UX person, get to know whoever manages analytics in your organization. You’re going to need that person, because you’re going to be asking them a whole bunch of questions. Second, the best way to learn is by doing. It’s about having practical questions that you want to answer. Say you went off and took a three-day workshop on how to use web analytics tools. If you don’t use that knowledge, it’s going to go away. Within a few weeks, it’s like you didn’t even take the course.
When you, as a UX person, are confronted with a question, look for opportunities to inform your answer with web analytics. Don’t be satisfied with asking the web analytics person. Learn how to answer that question yourself. If you practice that, over and over again, as you get used to the idea, you’ll start to learn about the data and its limitations.
MLl: Do you have any type of framework? In your book, you talk quite a bit about reports, but is there a certain framework that we need to get into, if you will, to take the first step?
MB: It’s about discerning what problem you’re trying to solve and coming up with a very concrete problem statement. After that, explore the data, try to get the answer, and then ask yourself if the question is actually answered. If not, start the cycle again. Sometimes you reformulate the question and dig for more data until the whole story is told. Other times you go until you run out of time and have to use the good enough answer, which in real life, happens all the time.
ML: In the book, you spend time talking about Google Analytics, which our company uses on an hourly basis. What questions should we be asking ourselves when using GA for user experience measurement?
MB: Goals. You care about goals in Google analytics. It’s very important that we, as UX people, think concretely about what our business is trying to accomplish goal-wise and how the website supports those business goals. Think specifically about what particular metric you’re trying to move.
Beyond goals, get to know what your analytics tool has to offer. As user experience people, we care a lot about paths because they can help us uncover navigational problems. They may indicate kinds of potential problems or pages that we could look at more closely. Or if we found out that a bounce rate on a page was 95%, and there’s no reason that it should be 95%, we would explore it more. It’s good for problem identification in that way.
ML: In Chapter 10 of your book, you talk about personas. How do we really optimize our UX for personas when people are still able to see only one version of the website?
MB: It’s important to understand the idea of persona in a user experience context. From a UX perspective, the persona is very focused on behavioral aspects of users. We care about what people do—how one persona’s behaviors are different from another’s. The persona itself is a communicative tool for making sure that the entire team agrees on who it is that we’re trying to design for, so that we don’t all imagine our own version of the user.
You can inform UX personas with web analytics data by looking at what kinds of words they use when they’re searching on your site, what topics they’re interested in based on where they’re navigating to, what kind of common tasks they have, etc. From a UX perspective, you would not base a persona entirely off of web analytics data alone, but would merge other data sources as well to create the overall puzzle.
ML: Besides Google Analytics, do you recommend any other tools that might be ideal for user experience measurements? What does your technology stack look like?
MB: At ITHAKA, we use Adobe’s suite of analytics tools. In particular, their A/B testing tool, Target, is one that we use a lot. Also, another very interesting tool from a user experience perspective is Crazy Egg. It gives you a sense of where people are looking and where people are scrolling on a page.
I’m fascinated by the idea of tools focusing very specifically on A/B testing such as Optimizely. Google Analytics is a super powerful tool and it features A/B testing functionality, but it’s not as robust as you will find in other tools, or in the case of Optimizely. Also, I’m very fascinated by the idea of making it easy to make changes to a website and test opportunities like you see in Optimizely or Adobe Target. It makes it easier to manipulate chunks of the page on your own if you know a little HTML, and then test it before having to drag a developer into the mix.
ML: Let’s say there is a new person in a role and they’re interested in user analytics to do some usability or UX inspection. Is there a ten-point checklist of things for them to look at? What can be an easy win for an organization that has not done this before?
MB: As a starting point, I would take a look at what content or pages people are actually using. It may be surprising. Then, we start to get a sense of what people are actually doing on the site. Does that match what we think they are doing? That can be eye opening and can serve as a starting point for all sorts of activity.
Similarly, take a look at what people are searching for. Does that match the language that your organization uses? Are people searching for things that are actually part of your business or on your website?
The last thing is to start looking at the most common behavior in terms of page-to-page navigation. You can start to form a theory about why they’re doing what they’re doing on your site. In this way, web analytics, the core of usability inspection, helps to uncover areas where there may be problems, which you then inspect. You can try to find out what portion of people are moving from page A to page B. That number will point out potential problem areas and give you context around your inspection findings so that it becomes not just a matter of expert judgment, but putting real usage data around that finding.
ML: Last question. You’ve been in this field now for many years. Is there anything that you wish you knew when you were getting started?
MB: My background, before I went to grad school, was that I majored in English and Music. So, I had a humanities background. For the longest time, I thought of myself as not particularly good at math. The thing that I didn’t understand was that working with quantitative data involves math, but it’s not the same thing. It’s not math in an abstract sense. You are solving practical problems. You’re dealing with very concrete, real world things, even if it’s website behavior. You’re still thinking about what people are clicking on. That’s not the same thing as learning calculus or algebra or something. I wish I had known that it’s not scary.
You can visit Michael’s personal website and check out his book, Practical Web Analytics for User Experience: How Analytics Can Help You Understand Your Users