Crafting Great Hypotheses
Make discoveries, not noise.
As part of our ongoing kick-off to UX testing in 2016, we’ve gone over how to form a UX research strategy by aligning company goals with product goals, and how to choose the right studies to form a UX research roadmap.
We’re taking it another step further and giving you insights and tips on how to craft hypotheses that elicit actionable insights from your studies. Actionable being the key word, since data without context isn’t an insight – it’s just more noise.Data without context isn’t an insight – it's just more noiseClick To Tweet
The benefit of hypothesizing
There are several benefits that come from forming hypotheses about your digital products. Namely, when you form a specific, testable hypothesis it forces you and your team to think through the assumptions in your designs and business decisions. Jeff Sauro, founding principal at MeasuringU, also says that:
“Hypothesis testing is at the heart of modern statistical thinking. Instead of approaching decisions with pure instinct and arguments in conference rooms, form a testable statement, invite users, define metrics, collect data and draw a conclusion.”
Define your objectives
Previously, we gave examples of questions that companies can ask in order to understand which studies they would need to include in their research roadmap. When it comes to defining objectives you want to cover within each study, it also helps to start asking questions. Generally these are Yes/No types of questions, such as:
- Do users understand your product and services?
- Does your business cover user / customer needs?
- Is your website easy to use?
If a user can’t understand the language on the site or complete a core task, congratulations – you now have an objective.
Craft a testable hypothesis
Crafting a great hypothesis requires three things: a testable assumption that can be answered by measuring impact, a specific variable that can be isolated and tested, and a prediction of the outcome. For example:
- Requiring the user to double enter an email will result in more valid email addresses
- Users will make more purchases if the checkout button is red instead of blue
- Users find items faster using mega menu navigation than standard drop-down navigation
Not only are all of these examples testable, there is also a specific variable that can be isolated and whose impact is measurable.
One of the hardest and most important parts of designing studies that elicit actionable insights is working with a good testable hypothesis. It takes time and practice to convert vague or even rhetorical business questions into testable hypotheses but the outcomes are worth the effort.
Like we said back in the first post in this series, success is when preparation meets opportunity. Research strategies, UX roadmaps, and hypotheses are all part of preparing, and therefor enabling, success. Join us next time as we take this series to its logical conclusion and give you advice from UserZoom’s experts for managing a usability study.
Phil got his degree in creative writing, where they told him he most likely wouldn’t be able to use his degree for his career. He obviously won that round. When not working with UX researchers he can be found teaching martial arts and working on his fiction novels.