How to create a UX research plan
Elizabeth Chesters takes you on a detailed journey of the DECIDE framework.
Research shapes everything in a business. The product, the target market, the future goals. And how we do our research defines everything, from success to failure.
But user research is so varied – there is no ‘one shoe fits all’ approach. The projects we work on as UX designers and researchers can be as varied as world news to sex toys, so it can be difficult to produce a template or model to work against. But that’s not to say it’s impossible!
For this ‘how to guide’ we’ll be using the DECIDE framework from Preece, Rogers and Sharp. We’ll then go through each stage of the framework to create a research strategy for a business. For the purposes of this example, I’ll be using a debt advice website.
Determine the goal of the research
The first thing we need to do is determine the purpose of the research. What do we want to gain from the effort we’re going to put in? What does the success of this project look like when we’re done? Narrowing down what’s in and out of scope for the project keeps the planning and project on track.
To start determining the goals we need to understand why this research project came about in the first place. Is it because the team is unsure how to design something? Is it because they have unanswered questions? Is the company concerned with laws coming into effect? The best way to do this is by talking to people on your team to discover how research would help.
For our debt advice website we could have the following end goals:
- Increase engagement rates of the content
- Uncover insightful behaviours of users in debt
- Reduce number of phone calls to the call centre
Keep these goals high-level and use the rest of the framework to narrow down. Also, there’s nothing wrong with iterating each section.
Explore the questions
We then need to explore what questions we need to ask to find the answers to our research goals. We can start by looking at what we know already, such as records of previous research or analytics.
Then we can start to see where the gaps are in our knowledge as a starting point for our questions.
In our scenario we could focus on asking:
- What are the mindsets of users when they look for advice on debt?
- What tools do users use to manage their debt and finances?
- What information do users seek when researching debt management?
- Which of this information is missing from our website?
- Why do users ring the call centre?
One of the most important questions you should be asking, regardless of the scenario is “Do I need to find the right design, or do I need to get the design right?” These two scenarios put into perspective the actual stage you’re at in your project.
If you need to find the right design, you need to explore different solutions. Here you should focus your questions on user mental models and observing behaviours.
If you have decided on a design, you then need to hone in on that design. Here you should focus on asking questions around the usability of that particular design, in order to refine it.
Use these same questions later to remind you what you need to ask users. All you have to do is lift these questions from your strategy plan straight into a discussion guide.
Also for context, add why you needed to ask these questions and who wanted to ask these questions. This will provide a reference if you forget the context at a later date. Future proof your own research strategy!
Choose the methods
At this point, we know why we’re doing research and the questions we need to ask. Now we need to explore how we’re going to ask these questions.
A good place to start is by looking at the types of questions you have. Exploratory questions like “What are the mindsets of users in debt” are best answered in face-to-face user interviews, to build empathy first-hand. Whereas questions that focus on uncovering trends or patterns are best answered in a quantitative way, like surveys.
In our scenario we could look at the following:
- Mindsets of users: 1-2-1 user interviews, field studies to people’s homes
- Tools to manage debt: surveys, desk research, ask in user interviews
- Missing content from website: usability sessions, poll on the website
Always look at the cheapest and fastest way to get your answers. You can plan research to the nth degree, but we don’t know what the research will uncover. If we did, we wouldn’t have to do the research in the first place!
Research needs to be approached with an open mind, with the expectation that it could change everything. Finding answers quickly will show you whether you’re asking the right questions or going down the right path.
Identify practical issues
We’ve now decided how we can uncover the answers to our questions. But all this is based on the assumption that everything will go according to plan. Wouldn’t that be lovely? Yes. Yes it would. But we work with people, the most unpredictable things known to man. So, next we have to identify what could possibly go wrong by looking at all our variables – like users, facilities, equipment, schedules, budgets and expertise.
- What type of users do we need to ask?
- Where do we find our users?
- What ethical issues do we have?
- Does the team have the expertise or knowledge to conduct the research?
- How much time do we have before the deadline?
Let’s look at some issues for our scenario more closely regarding the debt advice website.
- Where do we find users with debt problems?
- What ethical issues do we face in approaching users in a vulnerable situation?
- How much do we offer as an incentive to users?
- Which tools do we use to run a survey or a poll on the website?
- Where should we host user interviews?
- Do we expect users to travel to us (with limited funds)?
- How far would users have to travel?
- What types of moderators do we have on the team who is best to run the interviews?
- Do we have quite forthright people on the team?
- What biases do we have on the team?
- Are we all from highly educated and rich backgrounds with Macs?
As you can see, in just asking these questions we have quite a few issues to deal with. It’s fine to not deal with all these issues at the time of writing them. But it’s good to know what you’re going to come up against when you kick-off the project.
Attempting to solve these issues when you haven’t finished listing them all is futile. Partly because the next step is deciding how to deal with the issues. Eager beaver. But also, you can’t solve issues without knowing the bigger picture. You’ll end up chasing your own tail when inevitably finding related issues.
Decide how to deal with the issues
Now that we have the bigger picture of possible issues, you can decide on how to deal with them. For each issue, ensure you have back-ups and alternative solutions. Just because you thought of one way or experienced it before, it doesn’t mean it’s the best way to solve your problem.
In our scenario we shouldn’t be expecting users with financial troubles to be travelling to us. Especially anywhere that’s expensive. That means we’re going to have to find users, find out where they live and ask if they’re willing to invite us into their homes. As an alternative, we can make travel arrangements for them beforehand, so that they’re not out of pocket.
Let’s look at the other issue of finding users with debt problems. Arguably, people in these circumstances can be vulnerable. This could mean we have to choose to record and store data locally, rather than third party recording services like Lookback. At least this way we know that we don’t have to rely on third party security policies. Leaked data could have huge consequences for users and we cannot filter the details users share on video.
How to deal with the ethical issues
The original DECIDE framework specifically deals with ethical issues, which play a huge part in any research. As well as dealing with our research issues, we need to have a plan of action when dealing with users once they’re with us. Issues like informing users of what they’re signing up to and where we store collected user data.
In our situation, for the debt advice website, we need to look at the sensitive data we have to collect. How do we anonymise findings and how we could provide a copy of user details if a user requests it? We’ll also need to choose where we store the data we collect and let users know if we’re using third party services.
We then need to collate all this ethical information to invite users to participate, all while trying not to scare them away!
Evaluate, interpret and present the data
Finally, we come to the last part of the strategy. Evaluating, interpreting and presenting the data.
Here we need to look at how we use the wonderful UX insights we want to uncover. Will we be creating a new product? Are the insights to improve an existing system? Which stakeholders invested in the project and need the results?
In our situation we may need to use the findings in a number of business areas. Whether, adding user behaviour findings to acceptance criteria or tweaking your designs. Presenting back to the team and company in general is always a powerful way of presenting data. Sharing empathy with the wider team provides a purpose to the work we’re all doing and how we’re achieving our goal.
We also shouldn’t forget to document the research, from start to finish. Including details on the obstacles faced along the way, session materials and methodologies used. This way, the team can look back, see what happened and have a detailed starting point for the next research project.
Congratulations, we’ve just completed our research strategy and all we had to do was DECIDE!
- Determine the goals of the research; provide a purpose to the research
- Explore the questions; what needs to be uncovered
- Choose the methods; how to ask the questions
- Identify practical issues; obstacles to overcome along the way
- Deal with the (ethical) issues; dealing with the obstacles
- Evaluate, interpret and present the data; how to record the findings
Nothing says that what you do at the beginning of a project is set in stone. Iterate over each step, taking into account what you’ve learnt later on. If you highlight questions you need to ask after going through your issues, then add them to your list.
Unfortunately, there’s no one strategy to rule them all! But this frame at least provides a good place to start. So, why not get out there and try it before your next piece of research?
Main image by Sergey Zolkin.