What is included

The main method is simple and cheap: we give a person a task and watch in silence.

Discuss the scope

Scenario tests

We give a real task such as find and order this and observe without prompting.

Behaviour analysis

Where people stop, go back, get lost or abandon what they started.

Device testing

Different screens and browsers. What is obvious on a desktop may not fit on a phone.

Accessibility

Keyboard navigation, legibility and contrast. Useful for everyone, not only for people with impairments.

Forms

The narrowest point of any site: redundant fields, unclear errors and losing what was typed.

Report

Problems prioritised by impact on conversion, with an estimate of the effort to fix them.

How the work runs

A test with five participants takes a week including preparation and the report.

01

Preparation

We formulate the tasks for participants. A task must be about the goal, not about the interface.

02

Testing

Sessions with observation and recording, with no prompting from the observer.

03

Analysis

We consolidate the findings and separate individual quirks from systemic problems.

04

Report

A list of problems with priorities and possible solutions.

Five participants are enough to find most of the problems. This is a long-established pattern: the first few people stumble over the same things and after that the findings repeat. Two tests with five people each, with fixes in between, are cheaper than one with twenty.

Frequently asked questions

Ideally real customers or people from the target audience. At a push, employees from departments unconnected with the product will do. Those who took part in development are unsuitable: they know where to click.

Analytics shows that people leave at a certain step but not why. A user test answers the why: the person says out loud what they are looking for and what they do not understand. The methods complement each other.

You should. A test on a clickable prototype costs less and lets you change the structure while that is still cheap. After development, changing screen logic is far more expensive.

We will test the interface with real people

Tell us where your analytics shows users dropping off. We will run the sessions and show the causes.