The Web Platform
March 7–8, 2016: Training
March 8–10, 2016: Conference
San Francisco, CA

UX insights from a drunk guy

Austin Knight (HubSpot)
5:15pm–5:45pm Wednesday, 03/09/2016
Average rating: ****.
(4.17, 18 ratings)

Prerequisite knowledge

A foundational understanding of qualitative research and user testing will help attendees keep up with some of the terminology and concepts in the talk. However, this talk is designed to be digestible by the majority of the developer and designer community (regardless of experience or expertise), so anyone attending the conference should be able to keep up.


“Your site should be so simple that a drunk person could use it.” Will Dayble popularized this concept in 2013, but it wasn’t until a couple years later that it was truly put to the test when HubSpot quite literally adopted Will’s philosophy, namely by paying a guy to get drunk and use our site. That may sound like an entirely ridiculous proposition, and it is. I went into the experiment expecting that we would have some fun, generate some buzz, and put out a hilarious share-worthy article. For all intents and purposes, this is exactly what happened. But what I didn’t expect was to stumble across an assortment of legitimate, actionable insights into the ways that user research can and should be conducted. Despite the frivolous inception of our drunken user test, it turned out to serve as the perfect avenue for inquiry into what it means to conduct truly effective user research.

I discuss our unconventional experiment, the unexpected knowledge we gained as a result, and the techniques through which that knowledge can be applied in modern user testing. What can designers and researchers alike learn from inebriated users? How can they practically apply this knowledge to their own work? What are the struggles and pitfalls of traditional user-testing methods, and why do they exist? What are the best ways to overcome those issues and obtain honest user feedback? How can the benefits of a drunken usability test be replicated in more typical research environments? And most importantly, how can researchers most effectively verify the usability and simplicity of a design? I’ll answer all of these questions, leveraging an entertaining and lighthearted story to shine a light on a range of pragmatic research discoveries.

Through a case study of subsequent experiments conducted at HubSpot after the drunk user test, I’ll take a close look at the metrics that are directly impacted by user research and explore how they can be demonstrably altered depending on how the research is administered. I’ll list the myriad user research tools available today and discuss why, when, and how they should (or shouldn’t) be used. To wrap it up, I’ll discuss why it’s important for companies to have an openminded UX culture and how audience members can effectively implement those practices in the workplace.

Photo of Austin Knight

Austin Knight


Austin Knight is a UX designer, speaker, and author at HubSpot in Boston, MA. He currently oversees the UX for,, the HubSpot Blog, and a range of additional frontend web properties and tools, which are used by more than 4 million visitors per month. Previously, Austin worked with teams large and small in organizations ranging from startups to public tech companies. He serves as a UX and IxD mentor at Columbia University and General Assembly, speaks internationally on the topic of UX, and runs a popular blog, podcast, and newsletter at