How do you research?

Aside

This is a great blog post by Dave Jeyes at AddThis. It provides a great framework for the basis of user research.


Get Out of the Office

Modern technology makes it easier than ever to get out of the office without actually having to leave the building. People use AddThis tools for their blogs, news websites, ecommerce shops, and business sites. So we tried to connect with as many of them as possible––across the globe!

We asked for feedback through surveys, email, fireside chats, one-on-one video conferencing sessions, and independent beta tests so that people could participate however they felt most comfortable. This helped us bridge language gaps and get feedback from our global user community.

Ask the Right Questions

It’s important to start every research project with an understanding of what your goal is. Our initial conversations were around industry trends, website goals, and marketing budgets.

What really helped was when we asked people how to improve our products, as well as what “AddThis Pro” meant to them. This helped open up the dialogue on what premium features people really wanted, like the ability to hide pages from the content recommendation widgets.

We also wanted to learn what the right price was for AddThis Pro. Asking people about their willingness to pay can be fraught with issues, but we found the Van Westendorp model helped determine what is cheap versus expensive for a new product.

Get Real. Fast.

They say a picture’s worth a thousand words, and the same is true in user experience design. Showing people a high-fidelity wireframe, or better yet an actual web page, helps to communicate ideas more clearly and get better feedback.

The same is true for the fidelity of a user testing session. With tools like Google Hangouts becoming ubiquitous, it’s easy to put the user in control and watch how they interact with new features. This helped us learn that first-time setup is core to using AddThis Pro, and we started working to improve that workflow.

Be Ready to Change Directions

Creative work requires trying out different ideas, and the ability to kill off many of them so you’re able to solely focus on the right ones. This means being willing to try, fail, and change directions as quickly as possible.

When we tested out promotions for various with site visitors, we discovered that people really wanted premium AddThis widgets. This wasn’t in the initial plan, but helped to shape the direction of AddThis Pro.

Balance Feedback with Data

It’s important to balance qualitative feedback with quantitative data about how people are using your tools. Usually these two kinds of research reinforce each other, but there are times when they tell very different stories.

For example, some people thought that AddThis Pro could include more in-depth analytics, but the web traffic didn’t reinforce that notion. Instead we gave site owners a powerful new dashboard which has already surpassed the classic analytics in user engagement.

These are some of the learnings we garnered from testing and research for several months, and none of this would be possible without the participation and time of our user group. (So thank you all who took the time to give us feedback!)


 

  • Have you ever heard of user research before?
  • Have you ever used any of these approaches in your user research?

User Personas in Research

Aside

MailChimp is arguably the most accessible and fun to use email marketing company ever. Their practices have certainly influenced the way my company writes our software, and I’ve noticed myself using their blog as a serious source of inspiration …and if I’m really being honest, a reference guide.

Below is a re-blog of an excellent post on user personas.


 

New MailChimp: User Persona Research

A few weeks ago, MailChimp’s DesignLab posted images of our User Personas to their blog. As Jason explained there, we wanted to find out who really uses MailChimp. It was a question posed to us by data analyst Allison last year. We could broadly generalize about our users (savvy, self-reliant, techie, motivated), but we realized that we couldn’t rattle off the four or five archetypical MailChimp users.

What we needed was a clear idea of our current users, so we could better empathize with them, and in turn design for and delight them—especially with amassive redesign on the horizon. To reconcile who we think uses MailChimp with who really uses MailChimp, fellow researcher Steph and I embarked on a long-term study of our customers to learn who they are; what, why, and how often they send; what kinds of issues they face; where they work; and what kind of people they are. This helps us understand how MailChimp fits into their day-to-day lives, which in turn empowers us to design smarter.

So, how’d we do it?

 

Step 1: Interview MailChimp stakeholders to see who we assume our customers are

Our interviews with decision makers here at MailChimp were illuminating. When we asked, “Who do you think uses MailChimp?” nearly all of our subjects identified the same characteristics: smart, self-reliant, and techie. Steph and I decided to model an “ideal user,” Fred, after the aggregate data we collected from those interviews.

Fred is a great tool for a couple of reasons. First, he exposes our biases and assumptions. Second, he reminds us of the level of expertise we’d ideally like to see in our customers. He tells us to design MailChimp to empower our customers to communicate smartly and efficiently.

Step 2: Rank our pool of active users by industry

We took a close look at the industries people selected when they signed up for MailChimp. Turns out, nonprofits, education, and the arts represent a huge number of MailChimp customers. This was helpful, because we thought specifically about them as we redesigned the app. We could start to contextualize where folks will be when they use MailChimp, and in what capacity.

Step 3: Identify subjects from popular industries and interview, interview, interview

We took several of the top industries and started contacting users for in-person interviews. We met those customers at their workplaces—to present us with not only a human face, but a sense of the environment in which a MailChimp campaign is created. For instance, is the office quiet, or is there a lot of foot traffic? Is the computer a newer model or something outdated? What terms or phrases did our customers use to describe their work, their situations, and their emotional states? We ended up traveling to speak to customers in North America and Europe, focusing on Atlanta, Paris, London, and Madrid.

Step 4: Analyze what we saw and heard

After we visited folks at their offices, we organized and tagged our findings, and then looked for patterns. We discovered a lot of similarities across different roles or types of customers. For example, we had initially thought of our advertising agency customers as much different from our communications consultant customers; we think of agencies as big groups with lots of moving parts, and consultants as independent operators. But both sets of users manage multiple campaigns for many clients simultaneously, and thus use MailChimp in similar capacities.

At the same time, both our public relations and administrative assistant customer groups described themselves as too busy to learn all of MailChimp’s features; they hardly have time in their days to set up and send a campaign—much less learn and implement new functions.

Step 5: Share our findings with the team

When all was said and done, we ended up with five archetypical MailChimp personas: Fred, our “ideal” user based on our internal interviews; Andre, our developer persona; Eliza, our PR manager; Ada, the receptionist; and Mario, the studio consultant. These personas are meant to serve as guides as we design and develop MailChimp—who struggles with time and tasks? Who is quick to adopt advanced features?

We wanted to share these personas with the rest of the MailChimp crew in a way that’s easy to grasp at a glance. Our UX Director Aarron suggested we turn our personas into posters. Using tags from our interview analyses, Justin and Jasonfrom DesignLab went to work and created posters that now adorn the walls of MailChimp HQ. They’re hanging right by the espresso machine, where people from around the office congregate. Our hope is that the posters get employees talking about our users and their needs over a cup of coffee.

Some of the descriptions we used on the posters raised a few eyebrows around the office. “Isn’t ‘inefficient’ a criticism?” “What’s up with ‘profiteer?’” We strived for honesty in creating our descriptions. Profiteer is not a judgement—we have users who want to make buckets of money. How can we help? Likewise, inefficient is not a criticism—we have customers who need more hours in the day and acknowledge they could be working smarter. What can we do for them?

We realize that personas aren’t representative of all MailChimp customers; instead, we think of these personas as a snapshot in time of common users, knowing that their shelf-life is limited. In the short term, these personas were a big influence on our redesign process: we thought more about how teams collaborate and how individuals work in a multi-screen world. The personas also influenced our UX research team. Our five personas are a good start, but sometimes they could be the same person on different days, or there might be enough space to warrant additional, feature-specific personas to add a bit of nuance. They’ll help guide us as we consider new features and functionality for MailChimp users. Now, instead of wondering who really uses MailChimp, we can ask more pointedly, “Who would use this feature?”

 

 

 


 

  • Do you identify with any of the personas MailChimp developed?