We put your product or service in front of real users, we watch them use it, we check whether it meets their real needs and we assess how simple it is to use. It's the basis of user-centred design.

To improve an existing product by identifying the blockers, feed your teams with field input, validate a product before putting it on the market, and cut down on post-production fixes.

Both, depending on your context. Video-call interviews let us exchange live and reach scattered profiles; in person helps us take in non-verbal cues, hesitations and behaviours. Either way, every session follows the same thread.

User testing: putting your product in front of real users

Q: How many participants do you need?

For qualitative testing, a handful of participants per profile is enough to surface most of the blockers, no need to test dozens. Beyond that, you mostly see the same problems. If you're after representative figures, that's a quantitative approach, which we scope together.

You design for your users, but are you sure what they actually do with it?

A user test puts your product or service in front of real users. We check whether it meets their real needs, we assess how simple it is to use and its usability, observing real people use it rather than guessing what they do with it. You come away with field findings to decide on.

Let's talk about your project How we go about it

When testing makes sense

We often get called at a turning point in the project. It tends to look like one of these:

Checking that a product works before putting it in front of its users.
Verifying that a redesigned journey is genuinely clearer.
Settling conflicting internal opinions with field input.
Seeing which of the points flagged by an audit really get in users' way.

The takeaway

When the people who design a service know it inside out, their judgement is naturally biased when it comes to evaluating it. Testing with real users brings an outside view. And by crossing that feedback with an expert audit, we smooth out each side's biases, social desirability or the white-coat effect on the users' side, representation biases on the expert's side.

How we go about it

Prepare a protocol

We turn your research questions into an experimental protocol, with scenarios, questions to ask, prompts, and the metrics to analyse such as success rate, the path taken or the number of attempts. We always run a pilot test to make sure the instructions are clear and the prototype works.

Recruit the right profiles

When we identify precise criteria, we work with you or a specialised partner to source the right profiles. We plan an incentive matched to your users' profile and the time you'd like them to give you.

Run consistent sessions

Sessions run the same way for everyone, with a precise thread. We split each session into four moments: we welcome and put people at ease, we observe during the task while watching non-verbal cues, we debrief on the spot, and we stay attentive at the close, because it's sometimes in that off time that the user shares the most interesting thing.

Analyse and synthesise

We note the standout points on the spot after each test, then go back through the sessions one by one to extract the useful data and lay it flat. Each step of the test gets its own section in a synthesis document, with a short summary of the main points.

We've done it on real products

We run user tests in our research projects, on websites, applications and internal tools. For the Banque des Territoires, on a digital portal for teachers, we combined a usability audit with remote user testing. We interviewed 20 users: the tests confirmed the weaknesses spotted in the audit, such as the absence of a breadcrumb and unnecessary steps, and brought to light the frustrations tied to them. From this we drew 4 major blockers, prioritised to simplify the journey.

See our UX research case studies