The following is the transcript of an interview with Bryan Eisenberg.
Bryan Eisenberg co-founded marketing consulting firm Future Now with his brother Jeffrey in 1998. Bryan and Jeffrey also operate a blog at Grokdotcom.com. Future Now applies persuasion architecture to increase online and multichannel conversion rates so prospects purchase, subscribe, register, make referrals, or accomplish other goals that can be measured and optimized. Bryan helped invent and develop MAPSuite, a suite of software applications that allow nonexperts to apply persuasion architecture to their businesses. He is chairman and a founder of the Web Analytics Association and has authored several books, including the “New York Times,” “USA Today,” and “Wall Street Journal” bestseller Call to Action
Interview Transcript
Eric Enge: Can you provide us a brief background of yourself and Future Now?
Bryan Eisenberg: Future Now got started in 1998. At the time, I was working as a social worker, and before that as a teacher. And my brother decided that he wanted us to be in business together, because we didn’t have a great relationship growing up, and so it may seem strange to people now that we could finish each others’ sentences.
We built it based on one premise, which was, “people just aren’t doing things right”. Basic sales principles were not being followed. And I think it’s one of our fundamental differences from other businesses, most people come to the web either for technology or marketing. We come to it from a sales perspective. The question we try to answer is how do you actually get the transaction?
We built Future Now on the premise that sites weren’t converting very well, and it only took the market about ten years to catch up with us, but we are glad they finally did.
Eric Enge: What was your background prior to getting engaged with Future Now?
Bryan Eisenberg: I was working as a social worker and as a teacher, and my brother is an investment banker. But we shared one thing in common, which is we both had a passion trying to understand why people do the things they do. And before we actually went to Future Now, we were doing some marketing consulting and sales consulting for a few different companies. I’ve worked for a few dot coms, and basically helped them improve their sales.
Eric Enge: Can you talk a little bit about what happened with the formation of the WAA?
Bryan Eisenberg: We had been in many conversations about creating an association during the early Emetrics Summits, but no one had taken any initiative. So one day, I got on the phone with Andrew Edwards and we called Jim Sterne and said “Jim, we are going to start this thing. I will be chairman, you be president, and let’s start putting the pieces in place to get this thing started”. Then we got the legal structure in place and then we recruited the rest of the board and some of the key committee members. Now, we’ve grown to close to a thousand members already, all across the world. Pretty amazed at what we’ve accomplished in this time.
Eric Enge: Yes, it’s a great organization. Can you talk a little bit about the goals of the organization at this point?
Bryan Eisenberg: I think the primary goals are around education and getting people to understand the value of countable marketing and measurement. And, we’ve pushed hard with the development of courseware for the University of British Columbia. We’ve been involved in different issues around third-party cookies and standards around identifiable information. It is always going to be primarily around the same focus because there just aren’t enough people who really understand and know what it is that we do. And so, we just have to educate more people about it.
Eric Enge: And what do you think the big challenges are for the WAA at this point?
Bryan Eisenberg: I think it’s extending analytics past the pure analyst and reaching more of the business people, and showing them case studies about how valuable analytics is and why they should take an interest in it, and that it should be something that concerns the whole organization as opposed to just set up for the little analyst sitting in his desk in a corner somewhere. That’s going to be one of the biggest things that we are going to be facing over the next couple of years.
Eric Enge: You mean the process of getting the senior business people to understand the value of the analytics effort as an investment.
Let’s shift course a little bit. I would like to begin talking about what you mean by “persuasion architecture”.
Bryan Eisenberg: Well, persuasion architecture is a framework methodology and a toolset developed by Future Now Inc. They are basically tools to plan, build, and optimize persuasive systems online and offline, irrespective of the channel. So it could be television, radio; or it could be mobile. Doesn’t really matter. What allows us to do, is to build models based on all the assumptions we have about who are customers are, what action we would like them to take, and to leverage a deep understanding of their buying process and what’s going to move them along from early stages of the buying process all the way till full satisfaction.
Over the last couple of years is we’ve created an open-standard scenario language called PAXML, that basically allows you to tie in people to actions or events, however, you want to define it, to scenarios. Because an event or an action, doesn’t happen in a silo, doesn’t just happen because it’s on a web page or within a little web application. It happens because of what’s come before and what’s come after it and where the ultimate goal is. And so, we built this language to help model and understand that not every action is created equal, they are all different parts of the buying process, there are some that are much more critical than others. And at the heart of it is the customer, because they are in control.
Eric Enge: Right, that’s one of the big differences with the web environment that you’ve just called out, the notion that “the customer is in control”.
Bryan Eisenberg: Well, I think the web has basically just illuminated that. I think the customer always has some control, but in the past, they didn’t have as many options. Today, we have many more options because of the web. You can’t run away from a bad product anymore. It’s going to get out there and it is going to get out there quickly. And I think companies are going to have to start investing in real service and real products, as opposed to investing everything in their marketing and hoping that the marketing will overcome any shortcomings of their offerings.
Eric Enge: Right, There’s the old saying, you can run but you can’t hide.
Bryan Eisenberg: That’s exactly it. In advertising, all it will do is it will just accelerate the inevitable. But if the product is no good, you have to cope with the fact that today anybody with a browser, anybody with a camera, or anybody with a video camera is a publisher today.
With WordPress, TypePad, Blogger, Flickr, and YouTube anyone can become a publisher and they can get the message out about what’s happening with the product. The classic examples come to mind of the Dell exploding batteries, and how fast they did the recalls, one of the fastest recalls we’ve ever had in history. And I don’t think that would’ve happened if it wasn’t for YouTube and it obviously escalating all the way to MSNBC.
Eric Enge: It was a pretty compelling little video, wasn’t it?
Bryan Eisenberg: Absolutely, but even to the guy who tried to cancel his AOL account, and recorded the operator not letting him cancel it essentially, or the technicians from Comcast who were sleeping while doing an installation, really exposed weaknesses in these companies.
These are all classic examples of people being able to say, “hey no just wait a minute. This is not really right”. It is the reason that reviews have gotten so valuable in the last couple of years, more than ever before, and video reviews are even better. In the past marketers have advocated their responsibility of providing good information and being transparent, instead, they’ve just tried to catch hype, I think the Super Bowl commercials exemplify it all the time. Let’s just be entertaining and capture attention, but now there’d better be something behind it.
Eric Enge: Right. Is there an aspect of this which relates to getting the right users to your site?
Bryan Eisenberg: Well, certainly. If you understand your users, a big part of it is actually planning out what content they need to move them along, along the buying process. What are the trigger keywords, what are the value words that they use to search for a product like yours? And of course, as you develop rich experiences for them, more traffic naturally gets drawn in. It just happens to be one of the byproducts of persuasion architecture.
Jim Sterne once said, that once you see the persuasion architecture framework, and toolset, and methodology; you can’t imagine building a site any other way.
Eric Enge: Right. So then, it seems like another part is, you get the right user to your site and keeping in mind that the user is very powerful in today’s environment, has a lot of choices, going through a process of displaying or building your credibility with that user, is that fair to say?
Bryan Eisenberg: Oh absolutely! Roy Williams, our mentor always says that “speak to the dog in the language of the dog about what’s in the heart of the dog”. And if you spend the time focusing on using the customer’s language, and adjusting to their needs, and answering their questions; what you are essentially doing is building confidence that you’re a place that they can trust and that they can get value from. And, it builds tremendous amounts of credibility.
Eric Enge: Right, and then having done that the next step is just to make it easy for them to do what they came to do.
Bryan Eisenberg: Exactly!
Eric Enge: You also talked a bit about mapping customer personas and creating channels. So you can’t make the assumption that you have a single type of customer who may buy from you, you may have four or five types of customers.
Bryan Eisenberg: Exactly, there is no average type of customer.
Eric Enge: Can you talk about the customer mapping experience a bit more?
Bryan Eisenberg: We tell people, even if you don’t go as deep as developing full personas, with a lot of detail, and a lot of research behind it, just to start with understanding their perspectives. I’ll give you two sample perspectives, someone wants to buy, and let’s say you and I are going to go buy diamond earrings for our significant others.
I may buy, very spontaneously and assertively; whereas I see something, I look at it, okay I’m bought. I don’t pay any attention to the details. Right, it’s somewhere within what my budget range was, I am not going to worry about the details.
Eric Enge: Right, in out, you’re done.
Bryan Eisenberg: Right, in out, I’m done; that’s just my nature. You on the other hand, again maybe planning to spend a thousand dollars or so on these earrings, and say to yourself “I want to know what kinds of stones they are using, what’s the post looks like, and have lots of questions” because you may be much more methodical in nature.
These are different perspectives of how people buy, and human beings have always been hardwired into emotional, logical, fast, and slow. Now you can take those out to a different level, and then you start to identify that the different types of people on your site – this person is in the early stages of the buying process and at this level of knowledge, and this person is in the middle stage of the buying process, and this person is late in the buying process. All of a sudden you start developing, a much richer understanding of what kinds of information your customers are going to need in order to buy from you.
Eric Enge: Right. One of the interesting things that strike me is that there is evidence out there that says that the stage that someone is in the buying process is reflected a large degree by the actual search keywords they use.
Bryan Eisenberg: Oh absolutely, keywords completely illuminate the intent the customers have. I mean I have always told people that you basically want to segment your keywords by where they are in the purchase process. There is a big difference between “buy diamond earrings” versus “diamond earring types”.
In addition, the person who is looking for the perfect diamond is not the same person who is typing in “learn about diamonds”.
One is much a more spontaneous or assertive personality and the other one is a much more methodical personality. They are going to buy very differently. You can’t just roll up the diamond for both of them, and you need to split them up.
Eric Enge: Right, so it sounds like that a lot of this can be categorized as standard human behavior patterns.
Bryan Eisenberg: Yes, with an acknowledgment that humans are somewhat chaotic. So, it’s hard to look at things at the individual level, because on an individual level we can change a lot. But as groups, our patterns are fairly predictable.
Eric Enge: Yes, so today I might be methodical and tomorrow I’ll be impulsive and quick.
Bryan Eisenberg: Right, that happens a lot of times as we are going through a purchase. We may be trying to decide if we want to do something, and we start doing research and all of the sudden we find something that just clicks in our head, and you make the purchase instantaneously. It’s because we were dealing with modes of behaviors as opposed to a stereotype of a person behaves in this way. But, you have to have a plan that models different patterns of behavior.
Eric Enge: Right. Does analytics software play a role in making all this work?
Bryan Eisenberg: Yeah absolutely, because all we are doing is building a model. If you can’t measure the model then you can’t optimize it. Well basically, what we are doing when we are building out personas is making assumptions about the customers and the way they behave. Hopefully, if we had enough research behind it, and a good understanding of the product and the marketplace, we’ll be much closer. But essentially, when someone fails to convert our philosophy is a Six Sigma-like philosophy that everybody should convert, not that everybody is going to convert, but everybody should convert in a perfect model. And what we’re saying is, that if someone doesn’t convert, it’s because of either one of two things.
Either there was a flaw in our assumptions or there are flaws in our execution. Now, in the execution part, part of what we do in persuasion architecture is we plan all the variations of what we are go want to test in the planning stages, even before anything is built. We may have a couple of dozen ideas in mind, but do I know which one is going to convert the best traffic? I can have some ideas, but the court of last resort really is your customer.
What we do is try different variations of things, a different version of the headline, or a different version of a form, or whatever, and then we run through to the cycles using one of the multivariate testing tools essentially know which were the most effective experiences. So we do this very early on because everything has to have a reason, every word, every pixel, every page, every action has to have been planned for and have an measurable objective for existing.
Or else, you are just measuring implied intent and what I call accidental marketing, because you didn’t plan what the model behavior should be like, and any real scientist will tell you that if you really want to do scientific experiments, you have to understand the model and all the factors going into it. Instead, if you just pull up the site and pull up the analytics, can you define some customer intent? Sure if it’s really gross in terms. But when it’s scattered over a huge site like Overstock.com [the following is an illustration not using actual data] with tens with thousands of SKUs, there is no way to understand that data unless you started with a plan.
If you do that, you could have a situation where no one page has an exit number over 10%, but yet their whole overall conversion is not high, and you don’t know what to do.
So, you can’t optimize it, because there is not enough information from the analytics from at that point. Then people keep overly optimizing to the point where they can’t get anymore, and then they realize what’s happened is they have just fallen flat on the persuasion part of the site design.
Eric Enge: Right. You’ve got to have the plan in place to have the analytics set up from day one, and you have to have the personnel understand what they are doing with it to make it work right.
Bryan Eisenberg: Exactly.
Eric Enge: Excellent, so are there any special kinds of capabilities you need to have in the analytics software, or is it all fairly standard stuff?
Bryan Eisenberg: We are working with a lot of the vendors to adopt PAXML right into the software. We’ve been spending more time on understanding scenarios and what happens than just about anybody, as it has been our only focus in so many years. Several of the major vendors have agreed to integrate PAXML, and they are in different stages of doing that. But basically, there are two factors that we are looking for. We are looking to see how we measure linear scenarios as well as non-linear scenarios. Because we are looking for clusters of behavior, we don’t want to say that someone went from apple to banana to pear, sometimes someone goes from apple to pear, to banana, to pear, to apple; back to banana and that is still a conversion.
You can actually see the illustration of some scenarios if you visit: (http://www.futurenowinc.com/paxml) www.futurenowinc.com/paxml. If you go there, you’ll actually see the two different types of scenarios that we outline for people
Eric Enge: Right, so linear scenario is a step-by-step process.
Bryan Eisenberg: For example, a registration process. Whereas, a non-linear scenario could be the way we navigate when trying to decide which is the best DVD player or the best diamond earrings.
Eric Enge: An example might be someone who is wandering around Amazon looking for a book and they are not really quite sure what they want, and they try this or that category of item and they flip from here to there and then finally they pick something out later.
Bryan Eisenberg: Yes, but certain ones of those pages contain actions within those pages, and are more critical in helping us understand did we accomplish our goal of moving them along in the process or not.
If I want to get someone to download a whitepaper from our site, they may see the client success stories on the site, they may go to a little bit of our contact details, they may go through our bio-doc, and then they may go to the page that has this report that we want them to download. And then, they decide before they do that, they want to read our privacy policy and then they come back. Well, do I really care that they visited the privacy policy page? Ultimately, in terms of my measurement, no. It’s basically to point a resolution that satisfies them, and it doesn’t help me accomplish my goals. I am just trying to find the signal within all the noise and all the behavior.
And do I need to know that 0.3% of the people visited that privacy page? No, as long as they were able to get there and they were able to complete their goals, and if most of them follow that same pattern, then I know it was a successful scenario.
Eric Enge: Right. You can always just look at the privacy page separately just to make sure you don’t have 98% of the people bouncing off the site after that.
Bryan Eisenberg: Exactly.
Eric Enge: You don’t need to map that into your more focused analysis here.
Bryan Eisenberg: Right, if I see that the scenario is failing, then I’d go in and analyze and say, let me dig deep. And then, if I look at this point of resolution, I’d say 98% of the people are leaving from the privacy policy page, then I know I need to go in and actually dig deeper into that privacy policy page to figure out what’s going on.
Eric Enge: Right. So if I were to try to generalize what you’ve been saying, at least in the world of analytics, one of the big problems that I’ve seen out there is people not really knowing what they are trying to accomplish with their analytics strategy. The notion of a persuasion architecture is to get very focused on the customer experience and flow and the metrics that will help them measure and improve that.
Bryan Eisenberg: Yes. My basic premise is we are people. When we build a web site, we are building a machine, and the machine we build is one that brings in prospects, runs them through our sales process, and at the end of our sales process, we hope to produce or manufacture customers. And then eventually, they can come back and become repeat customers. Correct?
Eric Enge: Right.
Bryan Eisenberg: Did you ever see The Madness of King George?
Eric Enge: No, I didn’t.
Bryan Eisenberg: It’s a great movie if you really want to understand analytics. King George goes crazy, and the doctors are trying to figure out what’s wrong with him by studying what comes out of him.
Okay. Now, what’s happened is, he has input, right, stimulus, and then his body processes and there is a reaction, and then we are trying to understand what happened based on the reactions.
Instead of saying, hey wait a minute, maybe we should look at what’s going into the process, beforehand, then we’ll understand the outcomes much better. And what goes into the process, is your customers. So start planning your customers first, and saying “what do my customers want, what’s it going to take for them to buy from me”. I may understand my sales process, but what is my buying process? What do they need? And, that can also help you develop better products that are going to meet their needs. It’s way easier than trying to define reality based on what comes out of you.
It is the fundamental flaw most website owners are experiencing today, except but because of how bad things are on the web, we are still getting value out of it. But a lot of the people who come to us, are coming to us because they can’t get any further, that they start realizing that there has to be another system, and that’s where persuasion architecture comes up.
Eric Enge: So what else is involved in starting with the persuasion architecture approach?
Bryan Eisenberg: Really just one more thing. Training somebody usually takes a minimum of a week, all day, every day, and then several months to really get it right.
Eric Enge: That does it for today. Thanks Bryan!
Bryan Eisenberg: Thank you, Eric.
