Demonstrating the value of design improvements and other UX work can be done by calculating the return-on-investment (ROI). Usually you compare before/after measures of relevant metrics, but sometimes you have to convert a user metrics into a business-oriented KPI (key performance indicator).
Users’ “productivity” tasks differ from “engagement” tasks, in whether more or less is better for metrics like time on tasks, interactions, and page views. Such KPIs are important, but they must be evaluated relative to users' tasks.
A/B testing often focuses on incremental improvements to isolated parts of the user experience, leading to the risk of cumulatively poor experience that's worse than the sum of its parts.
Likert and semantic differential are instruments used to determine attitudes to products, services, and experiences, but depending on your situation, one may work better than the other.
Collect UX metrics to show how well your design is performing over time or relative to competitors. If numbers are down, you know what needs improvement. If up, ROI data is a key management tool.
It's important to study why users leave websites. Analytics tools give you two metrics for web pages: exit rate and bounce rate. Understanding the difference between these two numbers is essential for better UX design.
Analytics for websites or other UX design projects should drive the project forward to better business success. Metrics that make you feel good may not achieve this goal.
Tracked analytics metrics should be actionable: variations in a meaningful, relatively stable metric reflect change in the user experience. In contrast, vanity metrics appear impressive, but their fluctuations are not operational.
Conversions measure whether users take a desired action on your website, so they are a great metric for tracking design improvements (or lack of same). But non-UX factors can impact conversion rates, so beware.
A treemap is a complex, area-based data visualization for hierarchical data that can be hard to interpret precisely. In many cases, simpler visualizations such as bar charts are preferable.
The most desired user actions (macroconversions) may be too rare to generate enough analytics data for fast design iteration, so we can also analyze smaller user actions (microconversions) that are more frequent and are connected to bigger goals.
Like A/B testing, multivariate testing is a design optimization method that involves experimenting with live traffic to find the best impact on conversions.
If you’re working on digital products, you should be familiar with what statistical significance means in the context of UX research. Otherwise, your decisions may be based on meaningless numbers that could be due to pure chance and not a reliable difference between design options.
Analytics data are only as valuable as the insights derived from them. Some figures can stand on their own while others need further research to be interpreted. To use analytics data confidently and accurately, teams must understand the difference.
Numbers don't paint the full UX picture, so in the quest for conversion rate optimization, don’t lose sight of the fact that we’re designing for humans.
Analytics data can help supplement observations made during usability studies by providing evidence on the severity and generalizability of the issues observed.
