• Share
  • Share
  • Share
  • Share
  • Share

Posted by James Oliver on 30 April 2015

Our guide to the top 10 things you should consider before starting any type of user research

I’ve been learning about the importance of research and testing in the digital world for many years now and have been exposed to quite a long list of testing methods. These include creating multivariate a/b tests, observing participants through a 2-way mirror, scanning through page designs & watching their eyes on a tracker device (heatmap style), indepth interviews with people (before even building anything), card-sorting, group sketching/activity workshops, and on-the-spot guerrilla testing in coffee shops – wow that’s actually a lot of different techniques when I write them down! 

Photo of a notebook with a postit note saying 'Top 10' tips

A common factor in all of these testing methods is the need to plan and prepare properly.  It can be very easy to rush into a tried and tested or preferred research method, and progress too quickly without really considering your objectives or which technique is going to give you the best results. 

To help you avoid that pitfall I’ve put together a list of what you should consider before starting any research - this is intended for those with less experience or maybe just starting out in the world of user-research.

Top 10 tips:                                    

  1. Have a clear objective for doing any research
    Don’t just do research for the sake of doing research. What are you trying to find out?
  2. Talk to people who will be your intended audience
    This will give you the most accurate feedback and avoid wasting time. Ensure you prepare a set of screening questions that helps filter out people who are not the right fit for your session.
  3. Don’t break the bank! Work within your financial budgets
    Recruiting people can be time-consuming, expensive and a drain on resource. Consider other methods like remote-user testing (telephone/screen-sharing), on-the-spot guerrilla testing, services such as usertesting.com or online surveys. It all depends on what you need to find out. Sample size isn’t everything – of course it helps to have the biggest sample size possible but you must be realistic – speaking to 6 people is better than 0. We recently had great success with using social media as a way to recruit people, as opposed to hiring a recruitment firm or trying to spread the word through existing networks.
  4. Allow adequate time to prepare everything! 
    Remember it’s more than just ‘doing some testing’. Potentially you will need to prepare, recruit, facilitate, take notes, record, analyse, and create a report. If you can get assistance, it will really help to designate tasks. 
  5. Ask the right questions 
    You only need to talk about things you don’t know or understand. If the research and answers are already out there – maybe you don’t need to do this research in the first place?
  6. Always carry out a pilot test 
    One of the most common mistakes is to rush into research sessions without having sense-checked the questions or checked your running time. You can do the pilot test with anyone, better than not doing one at all.
  7. Avoid asking leading questions 
    It’s a very difficult skill and something I’ve found myself doing many times, but in order to get the most insight – you must avoid leading people or pointing to answers and finishing someone’s thought process for them, as you won’t get much value from the research otherwise.
  8. Do more listening and talk as little as possible 
    Keep your questioning short and concise – and if you need a little more detail on something just ask “why?” or “how does that make you feel?” (where appropriate)
  9. Record the sessions – video/photography/ Dictaphone 
    Always ask permission but try to record the session with video/photography and use a Dictaphone (iTalk is a free app on the iPhone that I use). This will likely come in useful for evidence, persona development or presentation snippets later down the line.
  10. Straight after the session, do a quick 5-min takeaway 
    If you have a long day of testing, this will help you avoid forgetting important highlights and start the analysis earlier with some quick high level insights.

Further Reading


line of people sitting on chairs with their heads not shown

Personas - providing a human face to User Experience data

Why Persona development is so important to the success of a digital strategy

3 Minutes

businessman sitting on a deckchair on decking overlooking a beach on a sunny day

Remote User Experience Testing - NUX Workshop Round Up

Our learnings from the recent NUX Remote User Experience Workshop

3 Minutes

pile of colourful CDs

What does IA mean?

Understand what IA (Information Architecture) is and the value it brings to a digital project

5 Minutes


If you think we might be a good fit for your business, fill out the form at the link below and we'll send you our agency playbook. It contains lots more information about us and what we're like to work with.

Send me the playbook