When you’re looking at how to measure UX, the metrics fall into two main categories: behavioural and attitudinal. As the idea is not only to spot problems and frustrations, but to address the issues with an overarching goal of continuous improvement, it’s a good idea to develop a range of KPIs to give you a full picture. So, what are some of the key metrics that will tell you where your UX is performing well or falling down?
Quantitative data doesn’t lie, so crunch the figures to detect the UX pain points that need ironing out. You can choose from a range of metrics to suit your platform, including...
How many customers fail to complete their purchase or convert to the desired outcome? This can be down to poor usability, resulting in frustration. For eCommerce sites, your shopping cart abandonment rate can be calculated by:
- Dividing the total number of completed purchases by the number of carts
- Subtracting this result from one and multiplying the answer by 100.
Average order value (AOV)
Improving the UX can increase AOV, so it makes sense to keep close tabs on what your AOV actually is. You can improve the AOV by prompting customers to buy more, for instance, by encouraging them to keep shopping once they’ve added an item to their basket.
Task Success Rate
The Task Success Rate (TSR) measures the number of correctly executed tasks and is ideal for situations where a task has a properly defined endpoint. While the TSR won’t be able to pinpoint the root of a specific problem, it serves as a very useful ‘red flag’ to spot where things are going wrong.
Ease of completion
Building out the picture of task success, you can use the seven-point scale Single Ease Question (SEQ) to understand how easy (or difficult) a task was to finish.
User error rate
The user error rate (UER) metric indicates how many times an individual makes a mistake, such as finding themselves on the wrong page of the website. A high UER is indicative that the UX may need work.
Time on Task
On average, how long does it take for a customer to complete a particular action, start to finish, in minutes and seconds? You can use this metric to discover what’s going on regarding user productivity.
Search versus navigation
Are your customers having to use the search function to find what they want because the navigation is inadequate or too confusing? Generally speaking, it’s better to design the navigatio to minimise the need to use search. Calculating the search/navigation ratio as an average percentage will give you a snapshot of how your site’s doing.
With attitudinal measures, you’re seeing to capture information on loyalty, usability, credibility and appearance. There are several different tools to choose from…
Customer satisfaction – CSAT
With this simple metric, individuals respond to the question: “How satisfied are you with...?” The result is given as a percentage and normally uses five possible ratings, ranging from ‘very dissatisfied’ to ‘very satisfied.’ Thanks to the speed of this metric from the user’s perspective, you can use it more than once, at different stages of the customer journey.
Net promoter score (NPS)
NPS uses a very simple question: “Would you recommend this company/product/service/experience or not?” While a recommendation is a great indicator of how you’re doing, it’s very important to put this in context, as a high NPS score might be an indicator of strong brand loyalty or personal experience as much as a thumbs-up on a great UX.
SUPRQ uses eight questions to assess the quality of the website UX, covering measures of usability, credibility, loyalty and appearance.
System usability scale (SUS)
The System Usability Scale (SUS) involves the completion of a 10-question survey, with five possible answers per question, spanning from ‘strongly agree’ to ‘strongly disagree.’ Users’ scores are averaged out to provide a benchmark – scores that fall below this acceptable threshold are a sign that your site needs work. SUS scores are given as a figure out of 100 but be clear that these are relative scales and not percentages when sharing the data.
Task performance indicator (TPI)
TPI is a management model based on customers’ top tasks, usually no more than 10 tasks. Task questions based on these actions are given to a sample of up to around 20 customers, who perform them in a testing environment. Based on how easily they complete the task, how quickly and the key problems, improvements can be recommended and the process repeated every 6-12 months.