We’ve found our UX metric scores have been invaluable for quantifying the impact that different design decisions have on user experience. They help shine light on key areas of improvement and direct our attention towards interesting user behaviors that can be revealed through deeper qualitative analysis.
Despite our success with using UX metrics in driving design decisions, we may have a problem: consistency. Our UX metrics are derived from a combination of tens of thousands of tests we’ve run with various tried-and-true research methods, as well as inspiration from popular user experience scores like SUS, NPS, and CES.
The inherent issue with this variety methods is that they each produce different looking outcomes: SUS produces a score from 0 to 100, NPS provides a score on a -100 to +100 scale, and CES can be reflected as a percentage or rating from 1 to 7. When we use multiple metrics to measure the performance of a design (as we often do, and recommend you do too), this makes it difficult to produce an overall design score based on the performance of those different metrics. The outcome is that our metrics are convincing as single data points, but together they create a smorgasbord of different numerical values that we’ve seen can be confusing through remote user testing and interviews with our customers.
For example, check out our Brand Analysis case study and how our research stack across Brand Score, Sentiment, and Loyalty appears:
How do you feel about these metric scores? Do they make sense? or does it feel more like a fruit salad with the wrong ingredients? We’re working on potential solutions right now, and we’d love to hear what you think!