Eighteen-And-A-Half Lamps

I have moved desks more times than I can count over the past few years at TechSmith. And with very open layout styles, you learn to appreciate the working condition preferences of those around you. Some like to mimic the feel of an underground cave, even going so far as to perch precariously on a chair to remove the fluorescent bulbs above their desk. Others aren’t happy until about 18-and-a-half lamps are turned on, blinding you as you try and chug your coffee to cope. 

As far as interfaces go, the subjective and polar draw to dark or light UIs seems no different. Our designers will regularly argue about it in any given meeting, tirelessly promoting one over the other. In terms of legibility and large amounts of text, typically it seems agreed upon that light UI prevails. For contrast and to provide focus, especially with video and media, dark UI is often helpful to take the focus off of the product and let the user concentrate on their content. Imagine if you had white bars blinding you as you tried to binge watch the latest Netflix series. Not cool. But is one inherently friendlier than the other? Is a dark UI automatically scary? I set out to uncover the worlds’ (okay so my…on this particular week) greatest mystery. 

Sometimes A/B tests feel like you’re asking a stranger which one of your children they like more… either answer is going to offend you because they’re both yours and you think they’re the greatest creations on earth. Others are comparable to asking a gal pal if you look fabulous in an ambitious outfit…there’s an expected outcome in mind and you’re looking for positive reinforcement, validation. But as bias as we may be towards one design or workflow than the other, we must remain silent and wait for the results. Let’s be honest, telling a developer or stakeholder that you feel like something is killer is just not the same as celebrating over a pile of cold, hard results. 

The goal: Discover whether we could design our iPad app with a dark UI and still maintain a friendly, approachable interface. 

The expected outcome (held innocently in my head): Hell yea the dark UI will still be approachable! It looks friendlier than a puppy and I love it. 

The game plan: Based on Microsoft's Product Reaction Card method for understanding product desirability, I had to pick five words that I hoped the survey participants would choose to describe the design. Half of the users would be shown the light UI, and half would be shown the dark. After, they would choose words to describe their first impressions. There were what felt like millions of adjectives to choose from, but the point is to overwhelm them with options to encourage gut instincts, instead of allowing them to ponder the sheet for days. The words I went with were approachable, friendly, easy to use, engaging, and intuitive, but there were a lot of similar words I would've also accepted as a “pass”.

The second half of the test involved actually having subjects try and use the UI to accomplish a task. It’s all well and good if they think the design is easy peasy, but if they can’t actually figure it out we’ve got a problem.

The results:

At least, that’s how it felt anyways…


There were stronger trends among the words chosen for the light UI, though words chosen for both were similar. But out of the thousands of possible terms…the participants managed to sync up with my word list for both the dark and light UI. Pretty incredible.

As for the usability, the darker UI surprisingly had a slightly stronger success rate. Our brave test rascals pulled it off. 

In conclusion, I am not saying that there is a right or wrong answer to the great “light vs. dark” debate. We made the switch to a dark UI for the iPad app for a number of reasons, not based solely off of this particular study. However, there is something very gratifying about a simple A/B test. And as far as my office space goes, I’ll stay as far away as I can from those eighteen-and-a-half lamps until at least 10:30 AM. #notamorningperson

Samantha NovakComment