Clarify the basics
Design research can feel overwhelming, even to people who’ve done it before! Clarify the basics with your team before you dive in.
Essentially, when we say design research, we mean the process of thoughtful exploration and learning with people to inform decisions. Design research explores possibilities, tests assumptions, and reduces risk by actively and systematically engaging with the world. It includes qualitative and quantitative methods, investigating tools and systems, and interacting with members of the public.
All 18F teams do design research. Design involves continuous decision-making, and those decisions are made better when they’re informed by people's perspectives. As a result, we’re committed to continuous research. Rather than seeing designed products or services themselves as the goal, we view the products or services we’re designing as the result of our continued effort to identify, understand, and address people's needs.
Because a collaborative approach increases the team’s overall empathy and efficiency, research is best done as a team activity. See our 18F blog on the impact of whole team collaboration on design research. This means the entire team, including your agency partners, shares responsibility for:
- Research design — Formulating a research plan, including research questions and interview guides
- Screening and recruiting — Determining who we want to learn from, and inviting them to participate
- Coordinating and scheduling — Managing research logistics (times, places, tools, people)
- Interviewing and moderating — Facilitating the interview or activity. The person who takes this on should be able to give their full attention to the participant(s) and should not be responsible for capturing responses
- Notetaking and recording — Documenting what happens during the research (for example, during a usability test)
- Observing — This is the default role for teammates who are new to research. Observers listen to interviews, observe usability tests, etc.
- Debriefing — Discussing individual sessions (usually immediately after each) to reach a shared understanding
- Synthesis — Finding patterns and themes across all research activities
- Reporting — Capturing and communicating findings from the research to partners and stakeholders.
On any given project you should only include the research activities that will inform the decisions you plan to make. Broadly speaking, 18F research falls into several categories. You can do research from any category on its own or in a sequence as you focus from problem definition to solution exploration:
Background or secondary research should be done on every project. This involves reviewing what is already known about the context you are working in or the people you want to learn about. Background research can save you and others time by reviewing findings from previous research studies or similar projects so you can build on what is already known, develop unique questions, and eliminate asking unnecessary, redundant, or triggering questions to people that could be harmful and reduce trust.
Background research helps you ask: what do we already know?
Foundational research is the research you do to identify and clarify the team’s objectives, assumptions, and constraints. On the 18F Methods site, discover methods help to build context for the problem you’re investigating. This includes; stakeholder and user interviews, heuristic evaluation, and hopes and fears workshops. Foundational research is primarily (though not exclusively) the domain of 18F Path Analysis engagements and results in, among other things, a problem statement as outlined on our GitHub repository.
Foundational research helps you ask: what problem are we trying to solve?
Descriptive research focuses on understanding, validating, and documenting the context of an assumed problem. On the 18F Methods site, decide methods help develop a deeper understanding of workflows and processes. It could look like describing the needs and characteristics of a particular audience like journey mapping or Mental modeling.
Descriptive research helps you ask: who or what are we solving this problem for?
Generative research helps you better frame the problem(s) you’re solving, spark new ideas, and reveal opportunities. On the 18F Methods site, make methods help create testable designs like prototyping and Wireframing to ensure your product reflects your users’ needs.
Generative research helps you better frame the problem(s) you’re solving, spark new ideas, and reveal opportunities.
Generative research helps you ask:
- What are our people's goals, behaviors, and pain points?
- What is their context?
- How might we address the problems we’ve identified?
- What does success look like?
- How do design opportunities impact the overall service and organizational context?
Evaluative research is the research you do to test assumptions, hypotheses, and the ease of use of design solutions. On the 18F Methods site, validate methods like multivariate testing, and usability testing help to understand how to determine if the product or service delivers as you designed.
Evaluative research helps you ask:
- Are we building the right thing — or if this research is done regularly, are we building the thing right?
- Does our product or service meet the needs of people who use it?
The following steps are repeated as necessary throughout each 18F engagement:
- Plan. Agree on the questions you want answered, the methods you’ll employ, how the team will contribute, etc. Decide who should participate.
- Do. Design and practice the research sessions. Correspond and engage with participants. Debrief after each session.
- Make research actionable. Review the data you’ve collected, look for patterns, write a summary of what you’ve learned, share it broadly, and delete recordings afterward.
As a distributed team, 18F defaults to remote-friendly ways of working. This gives everyone on our team—including our partners—a chance to participate, and lets us hear from people regardless of their location. We’ve written on our blog before about how to manage a research project and conduct moderated usability tests remotely, but our basic advice is:
- Try to conduct remote research via video conference. Find out what conferencing platforms your partners and participants can access.
- Use tools to emulate in-person collaboration. We use virtual whiteboards (like Mural) and collaborative word-processing applications (like Google Docs).
- Schedule regular co-working time. Use this time to work together on research plans, draft interview guides, or synthesize findings. Setting this up as a recurring event helps sustain the research effort.
With most projects you’ll conduct an in-person kickoff, during which you may have the chance to meet, interview, and conduct activities with stakeholders. If feasible, consider adding contextual inquiry as described in the 18F methods into your kick-off activities. Direct observation will provide firsthand understanding of your stakeholders’ and users’ processes and contexts.
We prefer in-person research when we want to better understand the environment in which users normally interact with a government service. We also prefer in-person research when our participants aren’t especially proficient with, or don’t have access to, video conferencing software.
Choose a research lead. A research lead is the team member ultimately responsible for setting the research agenda, explaining the team’s methods, tracking the team’s progress, and ensuring research quality. (While the research lead is normally 18F staff at the start, we work with our partners to identify someone on their side before the engagement ends.)
Discuss. Discuss your research plan and interview guides. Debrief after each session. Make sure to include partners in these discussions, which will go a long way in getting them on board with research findings as we discuss in our blog.
Determine research questions as well as interview questions. (They’re different.)
- Example research question: How do medical librarians find and make use of academic journal literature in answering questions for their patrons?
- Example interview question: Tell me about the last time you helped a patron answer a question that required you to review academic journal literature.
Be clear about who your research participants are. How we approach our research will depend on whether we are interacting with:
- Stakeholders (generally our agency partners)
- Users (those interacting with the system or product we’re investigating)
- The public at large
Note: Because some of our work involves creating solutions that are used by government employees, our users are not always the public. Avoid relying on stakeholders to speak on behalf of users, or expecting users to be aware of your stakeholders’ organizational constraints.
Make time for “research operations.” Continuous research requires ongoing attention to planning, participant recruiting, informing consent, managing data, and more, in addition to time spent on research itself.
Review government-specific considerations. Government-led research must account for legal, bias, ethical, and privacy considerations. Read these documents and engage your team in conversations about how you’ll account for them.