Paying for Better Design: Part 2
In my previous post, I explored the different benefits of embedding an agency into a client organization. In the second part of Paying for Better Designs, I will touch on the core of user experience design: usability testing. Like an agency embed, usability testing is a valuable tool for designers. However, clients sometimes find it difficult to see the value associated with the time and resources required. In this post, I address the flexibility of the process and the role it plays in the creation of a successful design.
User experience design doesn’t need to be sold anymore. A cursory Google search regarding the success of an application will be quick to surface entire articles dedicated to this discipline. Then why is it that usability testing is still seen as fat to be cut from the budget? At the heart of this notion is the prevalent myth that testing takes too much time and money.
Let’s unpack this by examining the different components of a typical round of usability testing.
Creating the Prototype
Prototypes can range from complex, high-fidelity designs within a fully interactive prototyping app to crudely drawn interfaces for what’s called paper prototyping. Either approach is valid. A product’s nature determines the type of prototype needed. For example, a complex application may not lend well to a static prototype. Conversely, users may be distracted by visual flourishes and gimmicky interactions when evaluating a simple flow.
A successful prototype focuses on the aspects that need validating, and does so in an effective way. Typically, I’ll spend a day or two constructing a prototype. In the earliest stages of product development, it’s important to create something quickly. We need to test with users in order to figure out the best direction to move forward with in our development sprints. As designs progress and we develop a firmer vision of the overall product, we engage in rapid prototyping and gain a better sense of the user.
Recruiting Users
Recruiting and speaking with users also requires varying degrees of time and complexity. A simple flow that is meant to be understood and adapted by a general, tech-savvy user base (for example, 18–30 year old low-income millennials) won’t require the same screening, incentivizing, and interview time as a tool targeted toward a different demographic (for example, 50–70 year old affluent neurosurgeons).
I like to follow the Nielson Norman Group’s recommendation of 5 users when conducting a usability study. Their findings suggest that “the main argument for small tests is simply return on investment: testing costs increase with each additional study participant, yet the number of findings quickly reaches the point of diminishing returns. There’s little additional benefit to running more than 5 people through the same study; ROI drops like a stone with a bigger [number].“
Conducting Tests
The length of the usability interview will vary depending on the complexity of the prototype, but I typically like to allocate 45 minutes. This gives the user enough time to go through the prototype while also allowing us enough time to explore any thought tangents that may arise.
There are a bevy of tools at our disposal when it comes to conducting user interviews which are valid means of offsetting costs. A quick tool for determining the efficacy of a page’s information hierarchy is fivesecondtest.com. With it, users are presented with a mockup for five seconds then asked to recall the information presented to them. Another tool is Usaura, a task-based usability test conducted online with users from around the world, which is an effective means of determining the ease of a particular user flow. However, no tools currently exist which can replace the value of an experienced designer sitting across from a user and asking the right questions.
In Conclusion: Just 10%
Repairing the damage done by a bad first version could be much more expensive than doing usability testing in the first place. A good example of this is the $300 million button case study published by usability evangelist Jared Spool. In it, he describes the amount of damage a poorly placed form had inflicted on sales for a major ecommerce site. The issue was that a registration form prevented users from purchasing items unless they had an account. This is now widely known to be bad practice, but at the time Spool conducted the usability study, it was commonplace.
After speaking with users and learning of their negative feelings towards being forced to sign up in order to make a purchase, the designers quickly changed the form to an optional entry at the end of the checkout. The results were a 45% increase in customer purchases. The extra purchases resulted in an extra $15 million the first month and an additional $300,000,000 within the first year.
The Internet is full of examples like these. A good rule of thumb is to allocate 10% of your design budget to usability testing. It’s an inexpensive method of ensuring that the other 90% is spent creating something that people will use.
Many larger agencies will outsource usability testing by contracting expensive third-party vendors to conduct usability tests within a formal lab setting. At Dom & Tom, we prefer to conduct our own tests. We find that we can ask better questions after having spoken directly with stakeholders and being the creators of the prototype being tested. Listening to the user’s feedback first hand ensures we’re available to ask the right questions, and immediately address users’ concerns. This process gives us the information we need to build the best possible product, and ultimately save our clients money over the long term.
