“User research” can be something of a loaded term, in our experience. To some clients (and industry peers), the only way to gather data is through large budgets and long timelines. This thinking often proves costly, as it can discourage people from undertaking what is, in reality, a simpler phase of work that still leads to well-crafted and relevant results.

Lightweight and targeted user research – lean user research – can bring heaps of value without blowing out a project budget. In fact, with careful planning, empathetic reasoning, and a thoughtful approach, it’s possible to gain more specific and insightful discoveries through ‘smaller’ and more intimate types of research.

The myth of more data

Research can be effective at a range of scales, from small focus groups in qualitative research to large population-spanning surveys in quantitative studies. However, usability experts Nielsen Norman Group (NNGroup) reckons that bigger isn’t always better. While scale testing can yield deeper insights and rarer issues, for your garden variety web project, a surprising amount of value can be gathered from just a handful of tests with small sample sizes.

As soon as you collect data from a single test user, your insights shoot up and statistically, you've already learned almost a third of all there is to know about the usability of the design. Critically, that transition from 'no research' to 'some research' offers the most proportional bang for buck - the difference between zero and even a little bit of data is astounding. As you add more and more users, you learn less and less because you will keep seeing the same things again and again.

Although you need “at least 15 users to discover all the usability problems”, NNGroup recommends testing with just five users over a couple of scenarios to optimise your results, your budget and your project timeline.

You can gain an immense amount of value from qualitative research across a small number of users when paired with expertise in a particular discipline. For example, an experienced designer will not only be able to plan useful research activities for the project in question, but they’ll have the necessary knowledge to both recognise patterns in the data and appreciate what they mean.

What lean user research looks like

As a small agency, the types of work we win don’t tend to call for extensive research activities. We’ve seen time and again the surprising value of small, well-designed user research activities within our discovery, planning, design and development framework.

Crib notes from our playbook:

2–3 short-run activities with a targeted crowd

By “targeted crowd”, we mean 5–10 people who have a deep connection to the audience groups they represent. This tends to involve people with lived experiences relevant to our end users, or project stakeholders and consultants with niche knowledge that can influence the quality of the final product. These could be workshops, interviews or specific exercises pointed to a given objective.

Card sorting, tree testing, and other IA research methods

IA testing methods are a surprisingly nimble way to validate some of the most critical aspects of a modern website, as with a relatively small sample size you can gather powerful insights to inform plans around navigation and content structure.

For most websites, information architecture activities typically involves around 10-15 people, again depending on the nature of the project and client. Depending on the scale of content offered by the product, service or organisation, that number can go lower or higher as needed, but any more than 20 would likely see diminishing returns. As a rule of thumb, if a project organisation includes multiple departments or independent areas of operation, that sample size should rise accordingly.

Rapid testing & concept validation with wireframes and prototypes

Leaning into NNGroup’s findings, we’d recommend testing design iterations over time with a very small, select group of users and stakeholders – between 3–7 people, depending on the nature of the project and client. The objective here isn’t to analyse what’s right and wrong with your design, but to improve and evolve your proposed solution according to what your target audience needs.

Small group interviews

Group interviews are an ideal way to dig deep into user needs, friction points, and potential pitfalls of the proposed design. With a smaller group, you can take your time exploring the less obvious (but equally pressing) concerns, and even uncover fringe use cases where inadequate designs still have a significant impact. While more time-intensive and subjective, this kind of deep, high-empathy research increases in importance the more mission critical your product is.

Surrogate testing

In this approach, internal stakeholders advocate for users, ideal for underfunded projects or projects where the target audience is for some reason inaccessible. For surrogate testing to be effective, the selected stakeholders must have in-depth knowledge of and a smooth communicative relationship with the target audience – basically, be familiar enough to speak on behalf of who we’re designing for.

Small research, big value

There’s so much to be gained from having direct contact with actual users, no matter how small or fleeting the engagement. You’ll find your assumptions challenged or your ideas validated, all for the cost of a few hours and a plate of sandwiches.

User research is best approached as a small, affordable and necessary stage, rather than an obstruction or a tick in the box. It reveals both the “unknown unknowns” and where stakeholders (and even us designers!) may be projecting our own expectations onto the user.

The earlier issues are addressed, the cheaper and faster they are to recover from. This is the real value of lean user research – the ability to move fast, learn quickly, and avoid unnecessarily breaking things. Research isn’t a luxury, it’s a scalable, tactical investment in the success of your product.