Enterprise UX: Complete Site Overhaul

Proposed design directions. Such a breath of fresh air from VSA.

Proposed design directions. Such a breath of fresh air from VSA.


The American Medical Association hired me to help them establish a user experience practice inside the organization. I made progress, the largest symbol of which is the complete overhaul of their current website, set for its first release in the fall.

You guys. When you hear people talk about healthcare being "behind" in terms of digital products and services, what you should do is change the word "behind" to the phrase "Great Scott Let's Start Over." I'm really not trying to be disparaging, just describing the magnitude of the problem.  And I wasn't even working in real healthcare.

Business Problem

(To make this easier to read, I'm going to write about this site as if it existed in the past. You can see the current site here until November.)

This site had three distinct main navigation menus, a taxonomy that became functionally extinct almost as soon as it was put in place, a wholly inaccurate search experience, and over 10,000 pages, 85% of which had gone unviewed for at least a year.

Tack on 20 disparate sub-sites, built by multiple vendor firms to compensate for all of these problems, which have shifted the digital brand to downright unrecognizable.

The content strategy had one bullet point: we should publish whatever it is people felt the AMA should be doing that day. No one wanted to set goals so that we could help measure the efficacy or impact of their content.

From a technology perspective, we had a full, heavy belt of single-use tools, including an over-customized CMS. Our 'apps,' which comprised everything from Become A Member to Contact Us required the involvement of three teams to make changes. The Contact Us app, a web form I described as "Please Don't Ever Contact Us," held a stunning four pages to send an email to one of five people in the AMA, all of whom worked for the same team.

The Big Picture

The place itself has been having an identity crisis for a while. The truth is, the American Medical Association is fighting for its relevance in either healthcare or medicine. They had a specific role for many years, but that role -- generally keeping the practice of medicine standardized -- has all but passed into regulatory and industry hands now. As a partial result, only 3-6% of practicing physicians are members.

There is much tug-of-war over this undefined future, with the website in the middle. It wasn't something we could iterate on to get to a better place, use current vendors for, black-box in its own corner of the world because it was just too big and impactful. We had to build a new foundation everyone knew about so that wherever the organization went, we would be able to have a consistent, clear way to put who the AMA decided to be out there. I really hope they take care of my very pretty, functional baby now that I've moved on. :)

But oh, the possibilities.


The C-suite dropped a deadline: We want a new site in a year.

I advocated for a long-term contractor and internal team combination and going agile, all the way, getting started right away. I could run that with the right budget and had the connections to make it happen.

But... 12 weeks later I was told that if we didn't have vendors people in the organization would not buy in or sign off on the direction, because employees weren't trusted with this kind of work. (Yes, that was a flag.) So we ended up with four vendors to do the work and a price tag to match. This is what I was told I must work with, so I trusted that they were right, and buckled in for the ride.

By now I had colleagues in content strategy, measurement and analytics and (finally!) development, so we set to work defining the new site with VSA, one of the best design agencies in Chicago. We spent the following six months doing all of the user experience and visual design work, firing two of the vendors who weren't delivering along the way.

(That part of the project alone could fill several blog posts on process.)


They haven't launched the site yet, but this is such a vast improvement and a blank slate for the future that the word "results" isn't really relevant.

A whole team of people

  • continually presented and iterated on design in rooms of more than 30 stakeholders for six months
  • overhauled the taxonomy and IA
  • whittled content down to roughly 600 pages
  • selected, refined and did the interaction design for a content-first design
  • reflowed existing external applications so that metrics could be gathered
  • helped the creative team with its image library organization
  • put together a new measurement strategy
  • planned out an A/B testing practice in place of usability testing


I directed, drove and influenced the enterprise site's entire user experience and visual design. I drove the creation of a true digital design and experience language that can be rolled out in all corners of the AMA, including its other sub-brands such as the JAMA (Journal of the American Medical Association).  And I collaborated closely with other enterprise technologists in development, analytics and content strategy to get it all done.



My Most Important Interaction Design

A few years ago I applied to be a business designer at IDEO. One of their prompts for the interview was to talk about the design I'm the most proud of, and I spent a long time considering what that was. Was it the app I designed for millionaires to make phone calls from their own phones on a plane? Mmmmmm... maybe not.

No, I had to go deeper. This is frickin' IDEO.

Software wasn't going to cut it, and I was starting to get less interested in that anyway. I'm way more interested in people, always have been.

So I pondered and sat and paced, until I saw it:

My absolute best, most intentional, committed design I've ever lived with was what I've created with my two kids.

Lemme get ahead of this one: I don't believe that you can "design" a family. I hear about designing everything these days -- it's a veiled word for controlling interactions. Design isn't about control.

I do believe, though, that you can set out to meet needs and improve things by being intentional about how you are in the world, and I am super intentional about what I'm doing with my kids.

I set up the piece the way you'd set up any presentation of user research and designs for its output: user input, user goals, results and feedback. I wrote about our evolution through early childhood, my divorce, all the way to two years ago.

I flipped through this piece to the end of it and saw this:

I started out an ambivalent, nervous mother, a drill sergeant marching toward the unachievable: maternal perfection. I didn't know that they were people too and that my job is to help them become themselves.
Today, my family is my best portfolio piece. My work, and theirs, demonstrates cognitive and emotional flexibility, a deep collaboration on and long term commitment to our quality of life, and the long-range vision necessary to consider what we want for our future as a unit.

I'm happy to report that today, at 16 and 14, we still do assessments, gather feedback, make adjustments, and change what we're doing to keep things running smoothly.

UX Lead: Tea Company Website


"We know the site isn't working and we know there's no reason for people to come back to it. We want a site that will engage our fans and new people with the brand."

Some of that is corporate-speak, but at the heart of it is a kernel of what I love, and that is a symptomatic patient: We are feeling the pain of this site not working and we need help from people who know how to figure that out. They recognize the need for change but don't know what to do.

A very rough concept of "what could be" on a homepage that changed based on what tea people drink at different times of day, and how weather might play into their decision.

A very rough concept of "what could be" on a homepage that changed based on what tea people drink at different times of day, and how weather might play into their decision.

On the other end of the phone, I did a little fist pump and thought, You're going to be SO glad I'm here. And even with the way this story goes, everyone involved is still glad they had me do the work.


Tea is like soccer in the United States. Everyone else plays soccer and drinks tea and those things have never become part of the culture in the US the way they have elsewhere. It could be just because we don't have the climate to grow it, as they do in Asia, and never had the empire the way other European countries did -- an accident of history? -- but in any case, tea has an oddly tough hill to climb in the US. People from other countries are often baffled by our love for coffee, which doesn't offer the variety in flavor and doesn't come close in terms of health benefits. (For the record, I drink both, but I do drink tea more now, a direct result of this project.)

This tea company started to change that, a few decades  ago. They made a name for themselves in herbal teas -- so much so that one of their tea names is used by tea drinkers to describe a whole category even if they don't drink that actual tea.

But tea culture has come a long way in the United States, and this company has some catching up to do.

They were looking at this problem using demographic profiles and aggregated attitude research, which were only going to get them so far. It's a comfort zone for a lot of consumer packaged goods companies -- when they say persona, they are talking about the mean and median of a huge group of people -- a person in a certain age and income range, whose hobbies include hiking and baseball, who drive Fords and shop at Target. But this type of information, while helpful to delineate a target market or what stores you might want your product in, does not tell you WHY people drink tea and therefore how to design for them. As one of my favorite people, Simon Sinek puts it, people don't follow a what. They buy into a WHY. If you know why people drink tea and why they choose what they do, you can enter into a conversation with them about why your tea might be just as appealing as that special-looking $4 silk pouch from Teavana.

Because It's About the conversation...

The conversation is the thing they were after. Brands who want to create that connection so successfully made by other companies have to recognize that you can't talk at people and tell them what they want any more, any more than I could ever make my kids eat or sleep (an early and difficult lesson).

...And You Can't Make Anyone Do Anything.

If you have something to sell it doesn't matter how much you like it or how you think about it. It doesn't even matter if you believe it will save the world. You have to understand people better, and persuade them to try your thing by making a product they actually need or want, which you learn from starting that conversation. And then you have to continue the conversation to build the loyalty other brands have.

Traditional ways of understanding "markets" creates the kind of conversation you forget as soon as it's over; designed  conversations create a deeper, more relevant conversation. Like when you meet someone at a party who believes hitting on you will result in inevitable 'success' if he lets you know how 'successful' he is. The person who will have success in meeting and connecting with people who want to be around them more is a better listener, a better connector, than that.


I recruited via Facebook and Twitter for just ... tea drinkers. No particular type or age. I wanted everyone who believed they were tea drinkers, and I didn't want only gourmet sophisticates or people who claim green tea saved their lives . I interviewed about 45 people, closer to 60 if you count friends and acquaintances who knew about the project (including people who don't drink tea) and transcribed the interviews. I asked opening questions like, "Tell me the story of tea in your life," or "Why do you like tea?" and that was it. Everything else was nudging deeper into the why. The interviews ended up ranging from 5 to 45 minutes, because sometimes you get people who don't really want to talk; in a recruiting post you can't really say, "Only sign up if you love talking."  

More than half of the participants made and drank a cup of hot tea while I talked to them.

From there I put all the results in spreadsheets and combed the results. Then I stared at those spreadsheets for two whole days before I came up with the four personas and mental models, which were strikingly similar once it came time to purchase, but for this: beliefs played a huge role in how people make decisions about what tea to drink, and therefore which teas they buy from where. (see gallery below for all that stuff). 

Normally you'd take those mental models, and with your client you'd fill out the bottom half to help everyone understand the ways the product currently supports the personas and what the client would like to support/be in the future. That way relevant ideas are created, discussed and agreed to, and the guesswork about "who will use this" doesn't have to happen, or at least, not as much because hey, this is people! You never know.

I was really excited about all of this and pitched it all with all the heart that was behind it. And in all my years of pitching, I've never had anyone respond the way an incredulous teenager might when you tell them they have to think about their future. They only half-grabbed my towrope. 

(To be perfectly fair to me, none of the work I did was sold into the original pitch or proposal, and it was barely mentioned at the client kickoff. So they had no idea what to expect and how it related to the project until a few days before I presented it. And they went through a persona exercise of their own in the project kickoff meeting that was... well, not going to help me.)

So, as you might have guessed, this is where the project goes a little sideways from my perspective. As they are wont to do.

Because it's not like they didn't listen, and it's not like I didn't set it all up in the meeting, but on top of not being prepared for what they were about to see, my work seemed "mushy" for them because it didn't come from "big data" (one of today's more misused buzzwords). They didn't see how it would lead to a website and what it had to do with design.  I explained it as best I could, in the plainest English I had available.

But I made the mistake of throwing a male into the personas.

Damn You, James.

Damn You, James.

Men don't drink tea, they said.

As incredulous as I was at the time, I just chalked it up to entrenched consumer packaged goods beliefs. It's ok. No big deal, I thought. But now I see that they were looking for something to point to that would indicate a larger flaw in my work, and therefore a way to set it aside and make the website they wanted to make.

A very human thing to do when someone brings in ways of thinking that are different from your own.

My persona, James the Enthusiast, helped me see that the hill to climb wasn't just for tea. It was for me, too.

It's ok. People don't have to take my work at face value or my recommendations. That doesn't diminish its value or quality, or even my job explaining it; it just means I have to shift gears in the conversation. Because how I work is challenging for people sometimes, and I can't just go off and make something based on what I believe.

That is called a hobby, not a job.

So after doing the best job I could selling a skeptical audience on the results of my research (see above for that point I made about how you can't make anyone do anything), I also wasn't able to talk to the client team directly about the mental models or the resulting feature sets. I worked with the team inside the agency responsible for the whole project, because there was trepidation around how it would go if we included an already skeptical client in a creative exercise they weren't comfortable with. I knew (and said) that this would make selling new features to the client harder, because they didn't come up with them or participate in the conversation. 

On the subject of getting my kids to try new foods, someone once told me that you have to expose a mammal to a new food five to seven times before they'll even try it because in the wild, it could easily be poisonous.


I am not one to swan about talking about how anyone who doesn't allow me to execute my vision just doesn't understand that I'm an ARTIST. I'm too pragmatic for that. And I'm not an artist. I'm a designer who gets paid not just to design, but to incorporate input that comes from that most unpredictable of animals -- a client.

A little idea I had to add music to the tea drinking experience using Spotify.

A little idea I had to add music to the tea drinking experience using Spotify.

And I had several small wins, which with this kind of work is what a person can expect on average. I had enough buy-in to create a navigation that is tailored to the personas I created, and  I was even able to help a small idea become an actual feature directly connected to what people do when they drink tea that might enhance an experience for them.  I was able to connect the dots for the client from my work to the site itself in a way that helped them buy in just a little bit more. Step by step. The site will be more usable for it.

Jury's out on the site itself. As of April my deliverables are mostly complete and it's time for me to move on. It might end up just being a reskinning of their current site; for now, most of my work might be forgotten or put aside, which would be a shame, because I know it's good work. But I've also noticed that what I do does have a tendency to live beyond the immediate reasons for doing it, once people have seen it.

That's because it is grounded in who people are, in all their brilliant difference and amazing similarity.

Also:  I'm drinking an amazing loose leaf Cream Assam right now from Todd & Holland.

Deliverables for this project:

  • Several Pitch Decks (I can't post)
  • Research Design & Plan
  • In-Depth Contextual interview transcripts and combing
  • Four Personas and Mental Models
  • Workshop for new feature set recommendations
  • Feature Set & Rough Concepts
  • Site Map & Navigation Model
  • All related internal deliverables for transitions to other teams