User research is the only real way to understand your customers, and make sure your content is meeting their needs.
However it can be time consuming and expensive, and you may not always have the budget to run large studies.
But the good news is that you can do research at low or even no cost. Your studies will need to be smaller and more focussed, but some research is always better than no research! And you can always build up your insights over time.
Define your goals
All research starts by defining your goals and thinking about questions you want answered.
Then, convert your goals into tasks or questions to ask in your research session.
Use data you already have
Look around your agency for data you already have about your customers and their needs. Call centre data is often a great source of information – what are your customers calling about? Can you create web content that answers their questions?
Similarly, have a look at correspondence data – do customers regularly write to your agency about particular topics or issues?
Does your communications and/or marketing team have any customer or audience data that you could use?
Desktop research
Desktop research, also sometimes called secondary research, can provide you with some insights and starting points to improve your content.
Have a look at how similar agencies in other jurisdictions do things, to see if there are techniques you could adopt or avoid. In commercial settings this is called a market or competitor scan.
You can also look for best practice guidance, especially if you’re targeting specific types of customers.
More reading:
Review your web analytics
If you’re looking at existing content, review the data available in Google Analytics. This will give you a good overview of how your customers are interacting with your content and can help uncover pain points.
See:
Review customer feedback from the Feedback Insights Dashboard
Every page on nsw.gov.au includes the ‘Thumbs Up Thumbs Down’ widget for customers to provide feedback.
As a first step, you can use Google Analytics to see how many positive and negative ratings your pages are getting. For a deeper dive, you can request a report of ratings and customer comments.
See:
Tips for no-budget testing
When you’re working with limited resources, keep the following points in mind to make the most of your no-budget studies:
- You don’t need a lot of participants – as few as five people can give you actionable insights, depending on what you’re trying to find out (more reading: Why You Only Need to Test with 5 Users - NN/g)
- Keep your tests short and very focussed
- Task-based and comprehension tests are likely to be more effective
- You can go low-tech – even paper prototypes can be a useful test tool with the right task
- Try and have a facilitator and at least one observer/scribe (two if you have enough people) – that way the facilitator can concentrate on speaking with the participant, while the observer captures the results
Recruiting participants
Recruiting participants can be one of the hardest parts of user research. A big advantage of engaging a specialist vendor to do research is their access to large pools of test participants, which you probably can’t match by yourself.
But there are some things you can do to find your own participants for small-scale studies.
Hallway testing
You probably work in an office with a bunch of colleagues – make use of them!
Put out a call through your internal channels for volunteers, or just randomly grab people in hallways, lunchrooms etc.
Make sure they’re from a different area of the business though – not subject matter experts, or from the digital team.
Personal networks
Your personal network can also be a source of participants – hit up your family and friends to do a quick test for you.
Guerilla testing
If you’re feeling brave, you can also try approaching people randomly in the street, or in shopping centres for example. Or if your agency is attending an event like a field day, local show day or industry event, tag along and ask visitors to the stall to take part.
Methods
Some research methods are better suited to small-scale, low-cost projects.
Usability testing
Usability testing typically involves giving a participant a scenario on a website and watching how easily (or not) they complete the tasks.
Example scenarios:
- Imagine you’re a high school leaver waiting on your HSC results, which have been released today. How would you go about finding them?
- You’ve decided to make your will and would like to make an appointment with the Trustee and Guardian to do so. How would you make the appointment?
In this type of testing, it’s important that you have the participants talk you through their thought processes out loud. You’ll need someone to facilitate, and at least one other person to scribe – take notes to capture the experience of your participants. If you’re running the sessions via Teams, consider recording the sessions to help with the analysis after completing all participant sessions.
Because usability testing is behavioural – watching what users do, not their opinions – you can get good results with a small number of people. Human behaviour is surprisingly consistent, and you will often start to see the same issues coming up with around 5-7 participants.
Usability testing your content is a good way to check if your calls to action and labels are clear and make sense to users. Look for points where participants are confused by the wording used or can’t find a clear path to complete the task. Listen out for signs of frustration or confusion.
You can run usability tests on your live pages – all you need is a laptop or a mobile phone. You could also run usability tests on draft, unpublished pages if you have a way of providing access to the unpublished content (like testing in person on your laptop).
This kind of testing can be done in person or remotely using Teams with screen sharing.
You can also do very low-tech usability testing using paper prototypes or printouts. This will work best for simple scenarios like ‘Where on this page would you click to [do something]’.
Further reading:
Card sorting – open or closed
Card sorting is a great tool when you’re working out how to structure and group your content.
Use card sorting when you have a bunch of topics to see how your users would typically categorise them.
An open card sort is where you just give your test subject the topics and ask them to group them and then name the group.
In a closed card sort, you provide the named categories and ask the subject to place the topics into them.
You can start with an open card sort to get an idea of your groups, then do a follow-up closed card sort with different participants to validate that you got the groups right.
You can also use card sorting to work out how to order content on a page – give your participants the headings and ask them to arrange them in priority order.
Card sorting can be a low-tech activity. In person, all you need is index cards or post it notes and a marker. For remote card sorting, use a whiteboarding app like Miro, Mural or Microsoft Whiteboard.
Because card sorting is looking to uncover your users’ mental models, which vary widely with life experience, demographics etc, you may need to include more participants – 10-15 people – before you start seeing similar outputs.
Further reading:
Surveys
Surveys can be a good way to get more detailed information from your participants, as you can ask multiple questions.
Surveys work best when you’re looking for people’s attitude towards your content, rather than their behaviour when accessing it.
A well-designed survey is a cost-effective and quick way to get a lot of data – you can send the survey to many more participants than you could include in other types of research. You can also consider launching a survey via a call to action link on your web page(s). The hard work then comes with analysing the resulting data.
There are a number of online tools that provide survey functionality, such as Microsoft Forms, Survey Monkey, and Menti.
Further reading:
Interviews
Interviews are typically the most time-consuming type of research, but they can provide rich, qualitative data. Interviews are best when you’re trying to uncover attitudes and opinions, or user needs and requirements ahead of developing web content.
The hardest part with this type of research is recruiting participants. You need to make sure you have a representative sample of your customer base, so need more participants than you do for usability testing.
Further reading:
Privacy, consent and data protection
Regardless of your recruitment strategy and testing methods, you must protect the privacy of your participants.
Participants’ names, email addresses, opinions and likenesses (if you’re recording remote sessions) are personal information and must therefore be managed in line with all applicable legislation and your agency’s privacy policies.
You must also ensure you record consent from your participants.
All records, documents, artefacts and data must be stored securely.
Verbatim quotes from your participants are a powerful way to show insights from your research, but make sure you anonymise any attributions in your reporting. For example, attribute a quote to ‘Participant 1’ or ‘Metro, age 35’ rather than using names.
Contact your agency’s privacy and/or legal team for more help in this area.
Further reading
If you’re interested in upskilling yourself in aspects of User Experience, testing and research, we recommend the Nielsen Norman Group website. Jakob Nielsen is widely considered the godfather of user experience, and the NN/g website provides a vast array of free articles and study guides, as well as paid training courses.
Some foundational articles to get you started:
- UX Basics: Study Guide - NN/g
- UX Research Cheat Sheet - NN/g
- When to Use Which User-Experience Research Methods - NN/g
Need any more help?
If you have any questions, or require assistance with anything mentioned in this article, submit a request via the webform.