Quick Tips for Picking Design Research Activities

Yellow carousel horses with another carnival ride in the background

If you find yourself with a pile of data you don’t know what to do with, or that fails to yield the insights you need, that might be because you chose the wrong method or activity.

Whether it’s a survey done to avoid interviews, a field study when the question calls for a quantitative survey, or a prototype test based on a fantasy hypothesis, it’s all-too-common to end up with a report that lands like a kitten at a dog show.

The problems that pop up at the end of the research process can only be fixed further upstream. So, unless you have a time machine, it pays to be extra intentional in your planning. This doesn’t even require much time, unless you have the bad luck to be facing a stakeholder who needs to be talked down off their pony. 

Here are a few things to think about to avoid the bad situation. These points might seem elementary, but organizations of all sizes trip up on this stuff. And if your research amounts to a lot of ticking the wrong boxes, then what are you basing your strategy on?

Questions before activities

Form questions. Gather data. Analyze data. 

Research methods are just ways to answer questions. You have to know your question before you choose your activity, and especially before you do your activity. 

This is the most popular mistake. Teams get this wrong all the time. They start with the activity and then think about their questions. We’re going to run a survey. What should we ask? We’re going to talk to customers. What do we need to know? Don’t do that.

Yes, sometimes your question is “What’s our question?” but if this isn’t your intention, it’s easy to end up reacting to what you find and wasting a lot of time. 

When you are clear on what you want to learn, you can choose the most effective and efficient way to get that information. And everyone involved in figuring out the research question will have a place in their brain to put the resulting insights. There is no one right method, but some are more or less useful and appropriate.

Separate methods from tools

It’s great that there are so many research tools and platforms now. It is also terrible.

Using software is not doing research. Any activity you choose, you should also know how to do with pencil and paper before you choose the tool to help you. Otherwise, you can’t choose the best tool, and you just might let how some software company makes money determine what and whether you learn.

OK, site analytics are really hard to capture with a pencil and paper, but you should be able to explain how gathering certain metrics answers your question and helps you make better decisions.

Reading is fundamental

Before you do any original research, give your team some time for background reading. Once you are clear on what you need to know, the entire internet is your research repository. You need an informed and critical perspective to judge the validity of people’s published studies, but there is so much out there…in books, articles, and even Reddit.

Don’t think you have to start from scratch, unless you are really getting into novel territory.

There’s a research question for you “To what extent, honestly, are we getting into truly novel territory?”

Quant or qual

If you need to understand something, you need to use a qualitative method—observation, interviews, diary studies. If you need to measure something, you need a quantitative method. And make sure it’s really quantitative, that is to say designed to yield statistically significant and generalizable results. 

Don’t try to use a quantitative method to answer a qualitative question just because someone wants “hard” data. Douglas Adams covered this in the 1970s, and I followed earlier this century.

Often you’ll need both, but you can’t measure something you don’t understand or that doesn’t really exist. Again, there are a lot of vendors ready to sell comforting, meaningless metrics. Caveat emptor.

Consider your category

Research questions generally fit into one of the following categories.  Once you’ve identified the category, you can use it to check whether the method you’re considering will give you what you need.

  1. Generative/Exploratory: How will this activity yield the kind of deep understanding we seek?

  2. Descriptive: How will this activity provide a detailed, truthful, and accurate picture of the situation or sequence of events?

  3. Evaluative: How will this activity enable a reliable assessment based on a clear set of criteria?

  4. Causal: How will this activity establish a cause and effect relationship, beyond simple correlation?

Cheaper is often better, but a harder sell

Getting time and budget to do research can be a challenge in organizations that have yet to embrace continuous learning. The good news is that clarity around your goals, priorities, and timeline for decision-making, makes learning fast, cheaper, and more effective. The standard for design research is the confidence and quality of the resulting decisions, not the polish of the intermediary artifacts.

However, pricing biases can create all sorts of other issues when it comes to choosing activities. More visibly expensive and technology-forward approaches may seem more effective to decision-makers than “just reading and talking to people” and then just talking about what you learned from talking to people. Again, this is something you need to fix on the front end by getting very clear on goals and questions before allowing the conversation to move to methods.

When you’re setting up remote research interviews, consider whether you even need video. You might, but if you don’t, you can save significant time and money dealing with the raw data.

Make time for skeptical reflection

We’re talking about a big box of tools, and a lot of different ways to use them. Do whatever helps you learn efficiently, when and how you need to. Just check in on a regular basis to make sure that comfortable habits haven’t replaced useful curiosity.

And if you need help with this, we have a workshop!

Previous
Previous

A Design Research Framework

Next
Next

You Can’t Do Enough