How to Conduct User Research and Build Features

“So, Megan, what do you do?”

What a loaded question, geeze. I do lots of things. I run. I eat. I hang out with my 5 rabbits (yeah, they’re awesome). Everyone asks me this question at every networking event, and I still don’t have a succinct, articulate answer. I usually reply with something along the lines of,

“I do user research and product strategy consulting for early stage startups.”

About 50% of the time, the inquiring mind will respond, “oh wow, that sounds really amazing,” clearly indicating that I’ve either convinced them that I am either smart or important, which is all that matters in Silicon Valley.

The other half of people will dig deeper: “oh, so what does your work look like then?” I like these people.

These conversations are usually fantastic. I get to expound upon some of my recent projects, go into details about what good product and UX look like at early stage startups, and hopefully do a little teaching about best practices. People particularly seem to love hearing my war stories about field research and synthesis.

What I find deeply concerning, however, is how few people understand the basics of user research prior to these discussions. There seems to be a huge gap in knowledge: how to recruit people to talk to, how to have conversations, what you should be talking to them about, how to synthesize the data, how to use this knowledge to make product decisions. Let me state this clearly and unequivocally: talking to potential users and understanding your markets at an early stage startup should be your top priority. This is understanding your users to figure out what to build, what to iterate on, and how to make money. This research isn’t rocket science, but a basic education in social science research methods certainly helps.

I’ve been doing this a while now, and what is sorely lacking is some concrete stories from the trenches to demonstrate the value of user research, especially in early stage startups. I’m guessing that’s why people are so excited when I tell them how I do my job. So that’s what I’m going to give you: a story, with exactly what I did, from beginning to…well, not end because this story is definitely still being written.


To get all meta now, my story is about a writing startup: TheRightMargin.

When I first started working with Shivani, the founder of TheRightMargin, she pitched it to me as a platform to help writers with writer’s block (read the origin story here). She had created it as something to solve her own problems, since she struggled to finish her own novels she worked on in her spare time. The key feature was dynamic, integrated content that helped you keep track of ideas, characters, outlines, maps, etc. while you write. The vision was to destroy Microsoft Word and make your life easier by stepping out of the way of your creativity — big, grandiose ideas.

But big, grandiose ideas need to start somewhere, which is what we’re doing right now by getting TheRightMargin off the ground. A writing platform can’t serve everyone who writes. It can’t be a platform with a billion features. A startup needs to start out targeted and narrow. I’m going to walk you through the research process I’ve helped TheRightMargin with over the last few months, which has led to a feature we’re testing.

Big Important Caveat: for the sake of the simplicity, I am outlining the process below as relatively straightforward because unlike a television show, I won’t win critical acclaim for confusing you with flashbacks and non-linear storytelling. The reality is, almost all of this happened in parallel and messily, with constant revisions. For example, before and after every single interview, I typically revise my interview guide. This is because my understanding of our users and the world is always changing. This is in contrast with your hardcore user interviews at larger companies or in academia, where methodological purism will often dictate that you stick to the original research design to maintain experimental integrity. I digress. Onto the actual “doing”!

Step 1. Brainstorm User Types

One of the big challenges we had at TheRightMargin is that the world of writing, even scoped to “long”, is quite wide and varied, so we could not design a product to meet specific needs without seriously narrowing our users down. Problem is, we hadn’t released a product yet, so we didn’t have any real users.

This is a common challenge for early stage startups and one that is perfectly surmountable. You brainstorm possible user types, do some pros and cons, narrow them down a bit, and start having lots of awesome conversations with people who fit your “possible user molds.”

Based on our existing understanding of the market, we guessed our biggest challenges would be getting people to shift off their existing established writing workflows and the diversity of processes. Initially, rather than brainstorm user archetypes, we actually started off discussing some important characteristics and demographics that might affect user behavior, since this could also help me develop my user interviews later on:

  • Organized/non-organized
  • Long vs. short form
  • Motivation for writing
  • Professional/on-the-side/students
  • Content: fiction, academic, technical, etc
  • Geography
  • Age
  • Gender
  • Technical ability
  • Language
  • Tools available and/or required to use
  • Autonomy and collaboration

From here, we were able to think about ideas for the types of writers we’d want to talk to — it was clear that some of the important variables we’d want to account for were:

  • Level of success — was someone a professional writer? Was writing just a hobby for them or did they hope to make writing their full time job?
  • Type of writing: what is the content of what they are producing? Fan fiction? Technical writing? Fiction? Short stories vs novels? How long were they spending on projects?
  • What tools did they currently use? What level of technical ability did they have?

To be successful in our user research, we’d want to talk to people who cut across these different areas and were not just coming from one of these areas. Since TheRightMargin had the potential to be a useful tool for a number of these groups, we’d want to narrow in and focus on a particular group and develop the tool for them.

Step 2. Create Data Collection Stuff

Before we did any actual research, I made sure we had the right resources. Templates, guides, and some basic organization always, always make your life easier. If you don’t deal with them in the beginning, you will regret not having them later.

For TheRightMargin, I created a “Customer Development” folder in our shared Google Drive to store all of our interview materials. There is a master interview notes template that we copy for each interview that gets retitled to the interviewees name (this should probably be something better). Here’s what gets recorded:

Name: (customer)
Age (estimate):
Researcher: (name)
Date:
Is the user signed up for the beta? Y/N
How did we acquire the user?
What communication have we had with the user before the interview? [Do we know them already? What do we know about their opinions of the product? etc]
Interview location/setting description: [Especially describe in detail if user typically writes there; particularly important for field interviews in customers’ workspaces, since you can understand their workflows and organizational processes. Pay attention to little things like their use of sticky notes, notebooks — anything that might look like a workaround OR a conscious rejection of technology. was this over Skype, Hangouts, etc? ]
Technology description, if applicable to in-person interview: [Mac or PC? iPhone or Android? Pencil/Paper? Sticky notes? Tablet, smartwatch, whiteboard…the list goes on. What technological (broadly defined) solutions does the customer use in their writing process?]
Interview notes: [Insert notes here]

The other key resource is my master interview guide, which also lives in the Customer Development folder. This is a living document and mostly consists of questions and topics that we’d like to cover in our interviews. I’ll have this open during interviews, but I almost never ask anything from it verbatim. It’s mostly something I use to remind myself what to cover.

The best user interviews are freeform, loose conversations driven by the user. Ideally, I’d never have to look at that interview guide and am just asking follow ups based on what the person is saying.

These user interviews are also entirely agnostic of TheRightMargin and are about understanding the person as a writer, their habits, their needs, and their struggles. Here are some example questions:

  • When was the last time you wrote? What did you write about?
  • Is there any information that you keep track of outside your actual writing documents?
  • What do you feel is your biggest challenge once you sit down to write?

Step 3. Recruit Users to Interview

Talk to any user researcher, and they will likely tell you that recruiting people to talk to is one of the most annoying parts of their job. There’s a reason that firms exist just to find people to participate in studies.

We initially struggled with how to find people to talk to, especially since we weren’t in touch with any writer’s groups. Some of our initial interviews were with acquaintances of Shivani’s who are writers, which is a fine first step, but has the issue of bias, since they are friends of the founder.

So, I did what any desperate researcher does when they’re starting out: I posted on Craigslist.

But, I know Craigslist has a serious issue of people who sign up for research sessions, regardless of qualifications, so I set up a filtering question. I mentioned we were looking to talk to writers, but I asked “What do you write?” I didn’t mention that we were looking for specific types of writers, namely people who write longer works. This meant I could specifically filter my responses. We offered a $40 Amazon gift card for people’s time.

I got some GREAT people from this posting.

Additionally, we thought that fan fiction writers might be a great group for us, so we posted on Fanfiction.net asking to talk to writers as well. Since it’s a tight knit community, there was some initial resistance, but one brave user was willing to give us a shot. Once she talked to us, she verified us to the rest of the community, and we got 2 more interviews.

Finally, at the end of every interview, you should always ask, “Do you have anyone else who you think would like to talk to me about their experiences?” and then do the same in the thank you email you send. We got a number of additional interviews through these referrals, since writers tend to know other writers.

Step 4. Synthesize Information

My biggest weakness is dealing with notes. I’d highly recommend setting up a weekly appointment to synthesize information from your user research — catch up on notes, read old research, synthesize data. Because this, at the end of the day, is probably the most important part of the process.

Initially, I had to both take notes and conduct interviews simultaneously, which sucks. If at all possible, I’d recommend having 2 people at every interview: 1 person to talk, 1 person to take notes. Christine joined TheRightMargin as a UX Designer recently and has been awesome to have along at the last batch of our research sessions. Thank you, Christine, for being AMAZING!

After each session, I like to do what I call a brain vomit for 5–10 minutes — essentially free writing all my impressions of what happened. Then, the next day, I’ll go back, clean up and incorporate everything so that other people can actually read the whole document. In reality, this doesn’t always happen, so the notes end up staying raw longer than they should. Like I said, a personal flaw of mine.

The way I synthesized the interviews from TheRightMargin is by pulling out pain points. I have a master pain point document in the Customer Development folder that has a heading for each person. I’ll have the interview notes open in one tab, and I’ll just go back and forth, adding to the pain points document. These are pain points, both explicit and implicit. It could be something they said like, “I hate find and replace in Microsoft Word.” Or it could be something that I’m extracting as a researcher like “Jane really seems to struggle with her self esteem and identity as a writer.” I also include everything that’s a pain point, however mundane, even something like “I want oreos when I write.”

I did this “pain extraction” process on the interview notes multiple times. I’ve probably read some of the notes upwards of twenty times. You really want to do this, so you become intimately familiar with your research and your market. It makes you more empathetic and thoughtful. You start processing these patterns in the shower, when you take walks, and in your sleep.

As it turns out, two really clear patterns began to emerge from this pain point exercise. One: writers, both successful and aspiring, realize that good habits are crucial to being a writer. In particular, writers really struggle to establish these habits. Two: writing is a very solitary, lonely profession and as a result, writers have deep morale issues. There’s a clear need to improve self-esteem and create connection.

So, TheRightMargin has decided to help writers with habits and morale. In particular, we’re going to help aspiring writers, the ones who really need our help.

Step 5: Brainstorm Features to Address Needs

TheRightMargin has a policy: they should only develop features that address one or both of the themes that we identified in the research. We really want to help writers, so it makes sense that the platform should hit on those two key pain points.

We quickly realized that while some of the organizational features in the platform would help writers feel better about making progress in their work and improve their morale, there was nothing in the platform to help them establish good habits. Enter, feature brainstorming!

What would be an MVP feature to test to see if we could help writers establish good habits? What does it even mean to establish good habits?

My research indicated that habits varied: almost all writers wanted to write regularly (with varied success), but some had word count goals, which others loathed. Some considered success to be based on finishing a scene, whereas others felt successful when they had just sat down to write.

Some ideas:

  • Email with personal stats
  • Email with inspirational quotes/resources
  • Word count -> total words -> deleted words -> graphs
  • Visual mapping of your story
  • “Jargon tree”

What did we actually decide to test? Freeform goals. When you login to write, we ask you to set a goal. You check off if you accomplish it, and the system keeps track of your accomplishments. There’s also an email reminder to ask you to check off your goal and set a new one.

Here’s what the initial whiteboard sketch looked like:

Here’s what it looks like now:

Step 6: Get Feedback

TheRightMargin has only recently opened up to users, so it’s in early stages yet, but the hope is that with feedback, the platform will be able to improve and be awesome for writers.

User interviews now include some usability testing and prototype reviews, so we can get specific feedback on the platform. But, we are still actively collecting behavioral data and will do so ad infinitum — it’s incredibly important to continue learning and narrowing our scope.

Speaking of which, are you a writer? We’d love to talk to you. We compensate with Amazon gift cards, and I’ve been told it’s actually pretty fun. If so, sign up here.


And this is why user research is critical. It directly helped identify an entire thrust of TheRightMargin’s product development that would not have existed without deep understanding and empathy for writers. Talking to people doesn’t hold you back, it empowers you. Don’t ignore user research, kids — it will help you innovate and do cool shit, I promise.

Show your support

Clapping shows how much you appreciated Megan Kierstead’s story.