Focus groups in user experience research: yay or nay?

All models are wrong, but some are useful. George Box
June 8, 2018 8:53 am

Focus groups in user experience research: yay or nay?


Within UX research, focus groups get a bad rap, often, this can be justified (for a number of reasons we will come onto exploring), however, this shaming is not always justified. Used for the correct reasons and facilitated in the correct manner, focus groups can become another useful tool in your methodological toolbox.

This article sets out to explore the pros and cons of focus groups within UX research, drawing on how and when we use focus groups here at SimpleUsability for context.

‘What we say and what we do are very different’

If you work in UX, this will be a phrase you will be very familiar with, and for a very good reason. It is well established and evidenced that as human beings we are not good at predicting our own behaviour for a number of reasons. This is one of the key drivers behind the stigma associated with focus groups, they are often used when they shouldn’t be, and when there is a more suitable methodology available, therefore the results may be misleading.

The role of a focus group within the product lifecycle is not about testing the UX or UI of a product, but understanding need-states of users and how the product can essentially alleviate those need states. When a focus group is used outside of these parameters we risk collecting unreliable data and therefore making the wrong design choices.

Therefore, focus groups should not be used for:

  • Usability testing (hopefully goes without saying)
  • Asking respondents to recall their experiences of certain products
  • Asking for direct feedback on proposed features
  • Asking for preference on prototype design choices

The ‘Gold standard’ of usability testing should always be observing natural behaviour, however, usability testing is only one aspect of a wider user research field.

All models are wrong, but some are useful. George Box

George Box tells us there is no perfect statistical model, similarly, there is no perfect qualitative methodology; different approaches work for different needs. Our job as researchers is to understand the research needs and create a study using the method (s) which will provide the best evidence to answer these questions. Here at SimpleUsability we live by the mantra of ‘Evidence beats opinion’ and therefore we will always build research to find the evidence you need. Observing interaction and user behaviour, such as usability testing, is often the most powerful approach. However, depending on the research needs and product stage, there are other useful and often complementary approaches to consider. This is particularly true when it comes to generative research early in the product lifecycle.

Generative research is key to kicking off a new project or development and should not be overlooked. If you don’t research until you have a prototype you are not benefiting from the full research tool-kit. One of the key principles of ‘lean design’ is to be able to throw work away and start again. Starting your project with generative research is another way of ‘hedging your bets’ and getting user input earlier in the product cycle and creating truly user centred designs.

When focus groups can be used:

  • Generative research
  • Early stages of new product development (NPD)
  • Understanding your users and the wider impact this has on how they may engage with your product
  • Early discovery research
  • Unmet need states (Precursor to prototyping or competitor analysis)

What makes a successful user research focus group?

As we’ve seen, we are not great at predicting our own behaviour and for that reason when we run focus groups here at SimpleUsability we ensure we are not asking users to predict their own behaviour or asking them direct questions. The art is in building the discussion guide to tease these needs out rather than asking users ‘What features do you want?’

This is achieved by understanding what the product/brand/service means to the user, when and how they engage with it, the reasons for engaging with it (if the product is already available). If not, it is more about understanding unmet need states and how a service can plug that gap.

It’s not about just asking the user ‘What features do you want?’, and then coming up with an extensive list. The moderator’s job is to decipher the user needs and explore how this fits into the product design. This leads to designs being built on evidence and not assumptions.

The UX practitioners here at SimpleUsability facilitate 1 on 1 user testing on a weekly, if not daily, basis and therefore have a very deep understanding of how users engage with products and services. Therefore when it comes to analysis stage and distilling a large amount of data collected, we are well positioned to filter the “It needs a big red flashy button” demands and focus on the information that is essential to generative research. Information such as: what is the user’s background with the product/brand/service, personal connection, what does that look like as a working relationship, how could the new product help grow this experience? In addition, focus groups allow you to develop a rich understanding of the language your users speak.

User research focus group checklist


  • Explore the wider aspects of your users
    • Environment, background etc. is equally as important in design thinking. Yes, we may not be best placed to articulate what we want and need but we can articulate our own beliefs and background
  • Use projective techniques to explore subconscious feelings and thoughts
    • Personification, word association, arguments for and against
  • Individual tasks to avoid bias
    • Content prioritization, understanding of language
  • Make the most of the group dynamic
    • Ask users to challenge each other and for counter arguments
  • Recommend further research in the product cycle to validate findings
    • Combination of indirect and direct observations
    • Real users directly interacting with the product early in the design cycle and periodically thereon out (1 on 1 in-depth interviews, competitor analysis, personas and hopefully usability testing)


  • Don’t ask for feedback on design choices
    • Feedback on prototype designs
    • Preference for language, icons, designs etc.
  • Don’t ask users direct questions or ask them to predict future behaviour (i.e. would you use this feature?)
  • Do not let the loudest opinion in the room lead others
  • Don’t shoehorn other research objectives into the research

The ‘ideal world’

In an ideal world, your research process should incorporate user research at the beginning and iteratively throughout the design cycle. Focus groups (designed and executed correctly) are simply another tool to allow you to tap into your users early in the development cycle. They are time and cost efficient and allow you to gain a first-hand glimpse into your users’ world to kick off your design cycle.