Before starting any new UX research project, there’s a simple question you must answer first: what do I want to learn?

It seems obvious. But how many research projects start with techniques — let’s interview customers, let’s do a diary study, let’s run a survey — before asking the most important question: what do I want to learn?

Starting from this point will give you a clearer set of project goals, and make it easy to choose the right techniques for your research.

This cheat sheet lists the top 10 things UX people want to know, and the best ways to collect data and generate insight for each one. Refer to it when you’re starting your next research project.


I want to know who my audience is

How to find out:

  • Review third party research — Ask your client or product manager for any audience segments or demographics they have, or search online for any third party research about the audience. Take the relevant parts from each report and add them to a set of user personas.
  • Run a contextual inquiry - Visit places where your audience will be and record what they do (e.g. car yard, supermarket, event). Document what you observe using a photo diarytime-lapse video or behavioural map.

I want to understand which content is most important to people

How to find out:

  • Analyse popular pages in web analytics — Check web analytics to find the most visited content. Create graphs to show the most popular sections (not just pages). 
  • Create heatmaps on the most influential pages — Another good way to communicate popular content is to superimpose click through percentages on page screenshots, or overlay a heat map. Tools such as  Crazy Egg or Hotjar will help you do this.
  • Ask your customers — As well as analysing the stats it’s good to run a survey to ask people what’s important to them, because sometimes the most visited content doesn’t always correlate to the most important. Create graphs to show the most popular sections based on survey results.


I want to understand the role of different touch points

How to find out:

  • Run a survey inside your product — Add a survey to existing websites/apps to ask people about the context they’re using it in. Graph the common uses of the existing touch points by frequency.
  • Analyse web analytics by device — Look at web analytics to see which devices are being used, and compare the behaviour across each device (e.g. page visits, important actions). Graph the different ways people are accessing the site/product by device.
  • Interview customers about their context — Find out when customers interact with different touch points in interviews — ask where they were, who they were with, what device they were using, the time of day/day of the week. Use a customer journey diagram to map out the results. Here's an example board that you can take a look at.

I want to know how people’s personal relationships affect their experience

How to find out:

  • Create a social network diagram — Ask about who else is involved with the experience in an interview. Create a social network diagram, with links explaining relationships.
  • Develop a behavioural map — Observe how people perform a task together in a physical environment (e.g. car shopping, attending events). Document what you observe using a behavioural map or photo diary.


I want to know how someone’s needs and experience change over time

How to find out:

  • Test in unexpected situations — Set a random alert/alarm for yourself or a research participant to use a product in unexpected situations. Document what you experience in a diary or update an existing customer journey diagram.
  • Run a survey at specific points in the customer journey — Add surveys that are tailored to each important stage in the current customer journey (e.g. signing up, cancelling membership, coming back after 6 months). Graph the most common reasons for interacting with the system at different points.
  • Create “day in the life” stories — Interview customers (or shadow them) and ask them to tell you the story of their experience with the current service — Signing up/installing, times they’ve used it, best/worst memories, why they stopped using it, etc. Write up the results as a “day in the life” or what ever time period makes sense based on the research.

I want to understand how people think about a topic

How to find out:

  • Make an affinity diagram — Conduct interviews with a focus on asking people to explain their thoughts and feelings towards a topic. Summarise the results using an affinity map diagram, and accompany each theme with direct quotes from the research.
  • Develop mental models — Ask people to draw/collage their experience with the topic or product/service. Create a mental model that aggregates the most common responses.

I want to learn about terminology

How to find out:

  • Run a card sorting exercise — Conduct a card sorting exercise with people using third party software to analyse groupings. Milanote is a flexible/easy option to do this in, or you could go for a more formal/structured option using Optimal Sort.
  • Test a draft taxonomy with customers — Setup a draft taxonomy/IA with a group of customers to test using third party software, such as Treejack. Analyse the results using the software, communicate in a summary report.
  • Make word clouds from user feedback — Describe the concept to a user and ask them to come up with a word to describe it. Create a word cloud to analyse common responses using Wordle.

I want to find out how well the current experience is performing

How to find out:

  • Facilitate user testing sessions — Run user testing of the current experience, either in-person sessions or using a remote service such as User Brain. Create a report with written or hand drawn recommendations, along with video snippets to illustrate common issues.
  • Try it yourself — Try the product yourself and keep a diary (written or visual). Consider using an unfamiliar computer/device which is closer to the average being used by your customers. Create your own customer journey diagram to describe your experience.
  • Map out the “pain points” — Check web analytics to find failure points. Pages with a lot of bounces, drop outs during important flows, etc. Add “pain points” to an existing customer journey diagram or validate any assumptions further with user testing.
  • Make word clouds using adjectives — Ask people in an interview or survey to describe the current experience in 3 adjectives. Create a word cloud to analyse common responses using Wordle.


I want to learn about competitors, or the industry as a whole

How to find out:

  • Test your competitor’s products — Run user testing of the competitor’s products. Create a user testing report that outlines the strengths and weaknesses of a competitor’s product compared with your own.
  • Make word clouds to describe your competitor’s products — Interview a competitor’s customers and ask people to describe the experience in 3 adjectives. Create a word cloud to analyse common responses. Compare responses to your own customers, and highlight any differences in the experiences that form these views.
  • Conduct a historical review — Review the long term history of the industry and how technology has impacted each stage in history 
    (e.g. bank statements over time). Draw a timeline of the changes at a macro level (i.e. 10 year blocks) and project how current trends might impact the industry in the next 10 years.

I want to test some early ideas we’ve had

How to find out:

  • Create paper prototypes to test with customers — Making paper prototypes is a quick and easy way to test early ideas. Present them to customers in interviews and write up the common themes, issues and opportunities in a report or presentation.
  • Use empathy tools to better understand customers — Empathy tools allow you to simulate different types of experiences (e.g. clouded glasses to simulate bad vision). Write up the results from the research in a report or presentation.

Getting started

To put this framework into action on your next research project, we recommend running a brainstorming session to go through each question with the project team to determine which parts are applicable to your project.

We use Milanote to run a session like this — see this example board to get started.


This is the first instalment of Milanote's User Research Masterclass Series. Part 2 will be coming soon.