|
|
|
|
Issue 20 // Relational Data
The Oakland A's likely haven't made many appearances in UX publications, so we're going to throw precedent to the wind and open this issue with them. The A's, as recounted first in Michael Lewis's best-selling book Moneyball and later in a movie adaptation of the same name, were pioneers in using statistical analysis to exploit market inefficiencies in the evaluation and signing of ballplayers.
Eventually other ball clubs caught on and caught up—to the point that now nearly every club devotes an entire team to this type of analysis. (The exception is the Philadelphia Phillies, who rent an analyst part-time. Look at their last couple of seasons and make of this what you will.)
So the A's are now employing their analyses in a different way: they're finding two players who—together—create a platoon where the sum is greater than their parts. This type of analysis takes patience, forethought, and a huge devotion to the research process.
And this is where we circle back to UX: skills become more valuable when shared, and data becomes more illuminating when you follow it to the ends of the earth. In this issue, Researcher Fernando Godina relays a story about putting our collective powers to the test, and fellow Researcher Gregg Bernstein puts forth his thoughts on radicalizing data. We conclude with links from the world of UX.
|
|
“… a successful business is the result of thousands of different elements that interact together at the same time.â€
|
|
Ideas and the projects they spawn don’t comfortably fit into a linear sequence. They pop up and take shape suddenly while other projects move forward in parallel. It takes a special kind of team to keep plates of many shapes and sizes spinning—a team full of generalists who can rove between projects and specialists who can go deep to make sure all details are considered.
MailChimp is made of people with diverse backgrounds. We’re self-taught; we’re-well educated. We’re biologists, advertisers, a rhetorician, industrial designers, painters, and poets. We come from all over the globe and speak dozens of languages. Our differences are our strength. Our diverse perspectives give us an array of knowledge and experience when we’re collaborating on different projects. We’re all deep-thinkers who enjoy trying to solve several problems at once. These are things I’ve always known, but a couple of weeks ago I saw first-hand how our diverse skills contribute to a collaborative workflow.
All hands were on deck as the UX team worked together on three different projects with looming deadlines. The research team was working on parallel studies that required sophisticated surveys and emails to thousands of customers. The design and development teams were busy crafting the next issue of this newsletter, and the research team was jumping in to help. We were hustling to get these projects done, but obstacles kept popping up that we couldn’t solve on our own. The only way we were going to meet our deadlines was to ask for help.
|
|
While working on a study of our Mandrill customers, fellow design researcher Laurissa Wolfram-Hvass and I struggled to figure out the right way to inject merge tags into an email we’d soon send. We were sending out a survey and needed the merge-tags to pass information, like customer email address and industry, from MailChimp into our SurveyMonkey data. But try as we might, we couldn’t get it to work quite right. Instead of pounding our heads on our desks in frustration, we let it go and asked some of our fellow researchers to lend a hand. They took a few moments to test the emails for us, and eventually the metadata we needed to pass from MailChimp over to SurveyMonkey was working.
We were just about ready to send our survey email to thousands of customers when we noticed some coding issues with our email template. Laurissa tried tackling the code herself and Aarron jumped in to help, but in the end we had to call in our pro-template coder, Fabio Carneiro, to help us sort things out. What was difficult for us was seconds of work for Fabio. We would have burned precious time had we continued to pound on a problem that was outside our area of expertise. By passing the baton to a colleague with the skills to solve the problem quickly and more effectively, we not only kept the project moving, we ended up with better results.
Meanwhile, back at the ranch, the rest of the UX team was working on the last issue of the UX Newsletter, but the articles needed edits before sending. Gregg and Aarron dove right in to refine the articles and smooth out a few bumpy narratives. Truth be told, we still shipped with one mistake, but hey, we’re human :).
All of this, plus another big survey project, were happening at the same time with every single member of the UX team contributing in their own way. Tasks moved fluidly between us as the clock ticked off the seconds left until our deadline. It made work easy, elegant, and fun. We have confidence in the abilities of our colleagues, so we know when a task is passed to a peer, it will get done. |
|
As we worked on three parallel projects that day, I saw two important skills that empower our team: communication and collaboration. Whenever someone had a question, he or she would IM other team members immediately to find an answer or walk up to their desks and start a conversation. In return, each question was met with an open and honest response and a willingness to help. Even while working on another task, colleagues made a point to respond in a few minutes, and if they didn’t have an answer they’d pass it on to someone who did. No matter the situation, there was never a dead end. We always kept projects moving forward.
Each member of the UX team has an area of expertise, but none is an island. Of course, this isn’t limited to our team. It happens throughout MailChimp, inside and between all teams. When it comes to making MailChimp better, we never hesitate to ask for help, and we're always willing to collaborate.
Our skills intersect as we work together, we ask questions, and we pass the baton. As a multidisciplinary team, we know we can tackle hairy problems concurrently, and that our individual weaknesses are mitigated by our collective strengths.
|
|
Our UX lead Aarron Walter wrote of this very topic in 2011, and we wanted to share how it applies to the MailChimp of 2014. For further reading, we recommend these articles: |
|
|
By the way, we're looking for a talented UI designer to join the MailChimp UX team. Is this you?
|
|
|
A popular corollary in design and writing posits that by emphasizing everything, we end up emphasizing nothing. We’ve all seen overzealous designs featuring too many typefaces in myriad weights that compete for our attention, or apps with far too many buttons for any one in particular to stand out as our logical choice. To get our point or intended actions across, focus and simplicity are key.
As a researcher, I’ve learned that only by flipping that corollary on its head does our team uncover meaningful and unique insights. We start with the premise that everything is important, and that every data point tells a story. By listening to each and every story, and then following those stories until they become epics, we achieve a mastery of our data sets and an ability to then focus, prioritize, and emphasize. We’ve radicalized our data, allowing it to dictate direction and scope.
|
|
The Journey is the Destination
|
|
Like most companies, MailChimp has an annual agenda that guides our work throughout the year. Typically this agenda is brief and amorphous—it’s not so much a directive as it is a few keywords that every team—from engineering to marketing to support—should keep in mind. A typical agenda could be about anything from the competitive landscape to nascent technologies.
This agenda, however, isn’t dogmatic. For instance, if we only look at the competitive landscape, we enter a loop of one-upmanship that fails our customers and our innovative spirit. By focusing on technology, we run the risk of shiny-object syndrome without asking how that technology actually benefits our customers.
At MailChimp, we employ a number of research channels—unsolicited email feedback, surveys, interviews, usability testing, and analytics, to name a few. By purposefully not making assumptions about what our research should uncover, we empty our minds of preconceived ideas and open ourselves to the data. It’s akin to the Grounded Theory method—only the data streams in to us all day, every day, decoupled from any specific project.
|
|
Unleash Columbo: Dumb Down to Smarten Up
|
|
To radicalize the data is to become subservient to it, no matter the form. By blanketly imposing importance on the data, we open ourselves to anything. In layman’s terms, we’re unleashing our inner Detective Columbo—we embrace befuddlement, we assume that everyone else is an expert and that every sentiment carries weight, and we trust that from this process the patterns and hierarchies will emerge later.
Every day, we receive emails from customers with suggestions and complaints about our MailChimp app. We read every single one of these emails, every single (work) day. We do this because we assume this data is important and that our customers know far more about using our app than we ever will. If we’re building for our customers, who better to learn from?
We take this same approach with our customer interviews. Once our data points us in a direction, like e-commerce, we visit customers to learn more about this topic and how it affects them individually. However, we don’t barge in and ask, “Can you tell me about e-commerce and how it relates to your use of MailChimp?†Instead, we take the long view and ask about typical and atypical days, how products are developed, or what brought them into their particular line of work.
By focusing on the narratives and tangents, we gain the insight that lends context to our original data sets. This is analogous to the Switch methodology espoused by The ReWired Group—customers hire products to perform a task, and our job as researchers is to uncover everything that influenced that choice. When we attribute expert status to our customers, we open ourselves up to empathizing with their situations and rationales.
Last year an email came through our app feedback channel that requested a “Notes†function within our Subscriber Profiles. It was the first and only time we saw this request, but we weren’t going to discount it outright—it was data, and therefore important.
We responded to our customer directly, learned of the use case, and this turned into an app development. The data we received was minimal, but by granting it legitimacy by default, we learned something new and added a feature with potential benefit for all customers.
|
|
Of course, following every data point as though each is canon has a logical endpoint. Therefore, we can complete our research mantra by claiming that each data point holds value until proven otherwise. Once we see enough evidence to contradict the data in question, we can move on.
For instance, as an email marketing platform we emphasize mailing list management and offer guidance about building a list. Through our feedback channels, we saw data that told us that unsubscribing from a mailing list is too easy via inadvertent clicks. The evidence mounted over time until it became a pattern and then a quick research project.
Once we looked over the data with a closer eye, however, we could determine that this wasn’t necessarily a data point worthy of further research and development—we examined who provided the data, the frequency, and the rationale. We could empathize with the feedback, but the use cases in question did not add up to a development that would improve our app—it would just appease a few and possibly upset even more. We followed the data and listened to the story until the story reached a conclusion.
|
|
When we research, we’re tasked with reporting our findings and providing supportive evidence. Depending on whether we’re a shop of one or a team of hundreds, we can all find value in a data-first approach to our work. To ensure that I allow the data to guide me, rather than me steering the data, I employ the following mindset:
— Everyone is an expert, and everyone is a researcher
The data that is shared with me is credible and valuable, and the opinions surrounding it hold merit, no matter the source.
— Every data point is valuable and worthy of collection
I will always collect, store, analyze, and verify. Over time, additional data and surrounding context will either transform the data into a story or discount its veracity. Either way, I didn’t let the data pass me by.
— Evidence and empathy, always
By radicalizing our data and allowing it to determine importance, every data point represents continuing education about our products and customers. With this principle firmly in mind, I can follow every lead and know that—even if proven unimportant—my research time was well spent.
|
|
Ask Us Anything
We want this newsletter to be a dialogue. If you have questions for the MailChimp UX team about how we collaborate, where we find inspiration, or even June's mad jewelry design skills, send them in! Seriously: hit reply and ask us anything. We'll try to answer every email and maybe even share our conversation in future newsletters.
|
|
|
|
|
|
|
