Democratizing research: lessons from a preschool teacher
As the demand for UX research grows, it’s not enough to simply keep hiring more and more researchers. In lieu of unlimited resources, we must figure out a way to democratize research. But what exactly does that mean? How can we empower teams to do their own research? Like teaching preschool, democratizing research is all about explaining things is a clear and concise way, starting out small, making sure everyone has a turn and establishing routines.
I didn’t start out my career at Microsoft, or even in tech for that matter. In fact, I started off my career as a preschool teacher.
Now you may be thinking, “what does that have to do with anything?” Well, you’d be surprised. Teaching preschool and scaling research have a surprising amount in common. Let’s start off by defining scaling research. What is it and why should you do it?
Scaling research: what is it? and why should you do it?
Scaling research at its core is about empowering teams to do research and providing tools and mechanisms.
Typically, people think of scaling research as a way solve for the issue of more and more study requests and not enough researchers. Since none of us have unlimited resources to keep hiring researchers, this work and expertise must be spread throughout an organization. None of our organizations are doing this organically, so we need to catalyze this shift toward everybody doing high-quality research all the time.
Listening to customers
Unsurprisingly, studies have shown that teams who talk to customers more often have better outcomes. Jared Spool and his team over at Center Centre/UIE have been researching what teams need to do to create great user experiences. They’ve found that one thing is crucially important above all else and that is exposure hours. Exposure hours are the number of hours a team observes users interacting with their designs, and it is the biggest predictor of success. There is a direct correlation between exposure hours and better outcomes. They found that teams who spent a minimum of two hours every six weeks observing and talking to users saw far greater improvements in their user experience than teams who didn’t. And teams with even more frequent exposure, had even better outcomes.
However it wasn’t just enough to have researchers observing and listening to customers — each team member had to be directly exposed to users themselves. Simply reading a research report or watching a video was not enough. Every team member needed to be directly involved in the research.
Now why is this?
Exposure to customers is an essential piece because it builds empathy and challenges assumptions. Seeing real people interact with the products and experiences we create reminds us that our users are actual human beings who are trying to accomplish real goals. This is why it’s so important that every on the team gets a turn at conducting and participating in research.
Behavior change
Sean Ellis, the author of Hacking Growth explains that conversion is a customer’s desire minus the friction they experience. In other words change = desire — friction. It’s not enough to just show your team the value of research- you also need to make it easy for them to engage.
By “friction,” I mean any little bump in the road. Anything that could slow down or prevent someone from doing research. Anything that causes just the tiniest bit of hesitate is friction.
Do your own research
To figure out what I needed to change, I did some research of my own. I started by interviewing managers and individual contributors about the current state of research to understand the biggest pain points and current work arounds. I also sent out a survey to my team to validate whether these signals I was hearing were real or just anecdotal. Here is what I found:
- 100% of respondents agreed that customer research was valuable
- 87.6% of the team thought that customer research was part of their job.
- Yet, only 14.1% thought they had the tools and mechanisms to conduct research.
I found that people wanted to do research, but it was too hard. Desire was already there, so I needed to minimize friction
Standard tactics
I got to work building out a toolkit.
I created a SharePoint site. I wrote a process guide. I crafted templates. I made diagrams.
And a few people did research. We increased the number of studies from zero to 29 in year (which is about 2 studies a month), and we had about 13 people participate in research.
However, I still wasn’t having the kind of impact I wanted. PMs, for the most part, were still not including me in their feature work.
For example, we launched a program called the Cloud Skills Challenge, where users could take a set of modules in a designated time frame, and if they completed them all ,they could earn a free certification exam. This program had a huge impact, and we saw engagement numbers soar, but at the end of the day, we weren’t really sure why it was so successful. Did people just want the free exam, or did they enjoy the competitive aspect? Was it something else entirely? Since we had not done research prior to launch, it was difficult to take the success from the Cloud Skills Challenge and apply it to other areas of the site.
Tools and templates do not create behavior change
I soon learned that just providing resources was not enough. Tools and templates alone do not create behavior change. If you really want to transform a team and catalyze change, you need to minimize friction and maximize desire in every step of the journey.
Where is the friction?
The first step in reducing friction is identifying where the friction is. One of the biggest pieces of feedback I got after year one was that people still didn’t know when to engage with the research program. Were we researching everything? All new features and content? What was the threshold for deciding whether or not we needed to do a study? When was it okay to move forward without research?
What people wanted was:
- A clear articulation of when and how to engage with research. What does that engagement look like? When does research become recommended and/or required?
- A decision framework for understanding how we prioritize what to do research on, assuming we can’t and don’t want to research every decision or every feature.
- A common set of research-related practices and competencies, with clearly defined resources that people could leverage to improve in those competencies.
In her book, Dare to Lead, Brené Brown talks about the idea of “painting done”. This concept is really about defining and sharing your expectations. It helps provide clarity and facilitates growth and learning. It’s important to have a clear idea of what success looks like once you’ve reduced this friction. Ultimately the feedback that I got was that the research program was providing resources, not driving consistency. Consistency is a sign of a mature program.
- People will feel like the program is really generating value when there are tools people can use to create consistency.
- Research is a way to make meaningful connections between business groups and get people aligned.
- Stakeholders will feel like the program is succeeding when others are citing things you’ve said and research is being used as a unifying force.
Give clear directions
My first step in reducing friction and increasing desire was to give VERY clear directions. Anyone who has spent time around children knows that you need to give clear and concise directions. No extra words. One task at a time. You do not tell a room full of preschoolers “we are going to make this project using a pattern.”
Instead you might say, “Let’s look at this flag. The colors go red, white, red, white. Hmmm, what color comes next? Yes, red! That’s right. And what comes after that? White? Okay, so the pattern is red, white, red, white.”
Similarly, you don’t tell a team “do research on things you think will have a big impact.” It’s important to focus on one task at a time and be explicit. Remember, these people may be experts in their field, but they are all new to research.
Establish clear guidelines
Humans are visual creatures. In fact, when it comes to the absorption of information, studies by 3M have shown that images and visuals are processed 60,000 times faster than text.
I made this decision tree to help illustrate when teams should reach out about research. Since we can’t research everything, I made a few very clear distinctions
- We do research on things that are new. If its not new for us, we likely either have already done research on something very similar, or there is other industry-wide research available.
2. We prioritize things that are customer-facing. If something is customer-facing, we probably want to do some research. If it’s an internal tool, it could probably benefit from some research, however, customer-facing things are prioritized
Identify intervention points
Similarly, I identified intervention points in the process where research should happen for the maximum effect.
There are three times when you are most likely to conduct research
- In the problem discovery phase, usually at the beginning of the development process, you’ll focus on identifying the problems your users are facing.
2. Next, in the solution phase, you’ll validate whether your solution is usable and addresses the correct problem. Usually, you’ll iterate a few times on it.
3. Lastly, once your solution is live, you’ll evaluate it and assess how people are using it, whether it meets people’s expectations, and how it’s performing from a KPI perspective.
Hold Hands
After establishing clear expectations, and outlining times when research can have the maximum effect. I turned my focus to how I could bring people along on the journey. Just like preschoolers, everyone needs a little hand holding when doing something new.
Just in time training
We established a self-service, consultative model. Essentially that means that our team helps PMs and content writers perform their own research and we offer help and consultation along the way.
We provide just in time training based on adult learning theory which states that:
● Adults prefer to learn from doing and enjoy hands-on experiences.
● They’re interested in learning something that has immediate relevance.
● They prefer to solve actual problems.
When team members reach out about doing research, we ask that they complete a few e-learnings before we grant them a UserTesting seat. Then, we work with them to apply the knowledge they have learned. In essence, we equip them with tools to solve their very real-world problems.
Grab and go templates
We provide a whole host of grab and go templates for every step of the process. Everything from the initial reach out, which is a research request, all the way through writing up a study plan, to writing up their results to share. We tested these templates and gathered feedback from folks as they were using them. This significantly helped us to reduce friction. It helped us identify where people were getting stuck, what was unclear and what was confusing. In the end, we drastically simplified these templates and provided examples to let people know “what good” looks like.
Scaffolding
Another technique that I borrowed from early childhood education is scaffolding. Scaffolding is a method of teaching that involves providing resources and support to students as they learn new concepts. As the students develop skills in those areas, the supports are gradually removed so the student can accomplish a task with no assistance. This concept was developed by psychologist Lev Vygotsky. He also coined the term “the zone of proximal development”. This theory states that there is a difference between what a person can learn on their own, as opposed to what can be learned with the assistance of another, more experienced person.
Start small
Similarly, we recommend that people who are new to research start small. We provide a lot of guidance and set up a series of meetings to provide help as people move throughout the process. Maybe their first study is just testing the current experience, or the experience of a competitor. Next, perhaps they test one feature or one interaction. As they become more and more comfortable doing research, they can tackle bigger and more complex problems. And, as the team becomes more comfortable doing research, the amount of researcher hours spent assisting employees decreases.
Establish routines
Routines are a great way to build confidence and help others know what to expect.
Routines build confidence
According to Dr. Peter Gorski of Harvard Medical School, knowing what to expect from daily life can develop a child’s confidence. Being able to rely on past experiences in a safe environment allows children to feel comfortable performing the same activities, such as putting on pajamas, on their own.
While routinely performing the same activities in a certain order may seem like just an organized way to get things done on time, this repetition also creates habits around healthy activities that children carry with them later on in life.
Make it part of the process
Similarly, we aim to create habits around doing research. There are many ways you can operationalize research and make it part of the process. In my experience, I went in to our roadmap and shared previous studies we had performed when they were available. I added a section in our spec template for customer research. This way, at the very beginning of the feature development process, PMs are already thinking about how and when to engage with research.
Be proactive. Reach out to people who are working on things that you think have a potential need for research. Everyone is busy, but if you try to make it easy for people to engage you’ll get a lot more done.
Lastly, set expectations for what the process looks like and how long it will take. On our research request form, and in all of our documentation, we note that research takes time. We explicitly say “the research process requires at least two weeks lead time. We will not be able to help you with a something that is launching tomorrow.”
Share
Lastly, I want to talk about sharing results. I’m sure that we all can agree that research is only useful if it’s shared. There’s a famous quote by Gregg Bernstein that says “Unseen research is wasted research.”
Sharing results is extremely powerful because it both reduces friction and increases desire.
I’ve found that the best way to show impact is to highlight stories of people who have had good experiences. For example, ask people who have performed studies to share their experience at a lunch and learn. Highlight one study every month in your monthly business review. Include these learnings in your state of the business. Plug into the channels that are available to you, and highlight the great work that’s being done. This will not only inspire others to participate, but will also provide recognition to those who are participating in research activities.
Another approach that has worked well, is to include different stakeholders from across the team in your studies. When there’s more people doing research you can bring in more perspectives. For example, when we interviewed our users about how and why they re-engage with our sites, we included PMs and content developers in the calls. During these sessions, they could ask much more nuanced questions than I ever could about migrating a system or building an app because they have deep technical knowledge of the space and therefor were able to understand what the participants’ issues were in a way that I couldn’t. Afterwards, when these folks returned back to their respective teams, they became ambassadors for the research program and in essence, force multipliers.
This is a great way to make connections across the team as well. The next time those people are in product discussions or deciding which strategy to go with, they will be reminded of the positive experience they had interacting with the research team and will be more likely to reach out in the future.
Results
I just covered a lot of information about different strategies I used to transform my research team here at Microsoft, but how do you know that these tactics worked?
At the beginning of the year, we set some goals for the program to track if the changes we were making were successful. Those goals were to:
● Increase in number of studies the team performed by 40%
● Have at least 60% of the product and design team complete training
● And to shoot for 75% of our features that are customer-facing involve research
After implementing these strategies from early childhood education, we increased the number of studies by 80% YoY, from about 2 studies a month to roughly 6 studies a month (which if you recall is inline with the recommended 2 hours at least ever 6 weeks).
We had 88% of the product team complete training
And we increased the number of customer-facing features that involved research by 561% in one year.
Most importantly, we significantly reduced the number of features that had to be reworked because we did research up front to inform our decisions. This saved time, development hours and built trust within the team.
Summary
In summary, reducing friction and increasing desire was what ultimately lead to the transformation of the research program on our team.
These basic concepts from my background in early childhood education had a massive impact on our team.
- Give clear directions. Be explicit and don’t leave room for interpretation. Pick a few easy to remember distinctions, and remember, these people may be experts in their field, but they are all new to research
- Everyone needs a little hand holding. Don’t just expect that by providing templates people will do research. You need to make it easy for them to engage and bring them along.
- Build routines. People, kids , adults- everyone loves routines. They help us to know what to expect and to gain confidence. Don’t forget that the sign of a mature program is consistency. Routines help everyone know what to do when and what comes next.
- Share. Your research is only useful if it’s shared.
Laura Klein has wonderful quote about the value of research; “Research, in itself, is not particularly valuable. It’s only the application of the research to product decisions that makes it valuable. Making the results of the research available to everybody means that, hopefully, more decisions will be made based on real user needs.”
More people caring about customer needs and issues means better products for everyone. Scaling research is the best way for us all to create better experiences.
