A Guide to UX Metrics

UX metrics — what are they?

UX metrics, or UX indicators, are quantitative data that allow us to determine, measure, and compare a website or mobile application user's attitude, reaction, and behavior.

Let's remember that the meaning of this term is suggested by the word indicator itself, which means, according to Merriam-Webster Dictionary, "one that indicates."  For example, conversion rate and bounce rate reveal problems with usability.

We can also learn about that from UX research (e.g., using focus groups, in-depth interviews, and qualitative data in general).

Colloquially, indicators help us translate characteristics, phenomena, processes, and their results into more understandable language.

We can express them more precisely, which, of course, is used to introduce changes.

Sales are easy to study and express. Human behavior and attitudes require a more creative approach and the discovery of where a particular characteristic manifests itself.

UX metrics can be divided into the following:

• Descriptive metrics, usability metrics that provide information about events.
• Perception metrics that are used to determine attitudes (such as engagement metrics).

Usability metrics help us find out how easy/difficult it is for users to perform a given task.

They allow us to see how much time a particular task requires and how often users make mistakes while performing it.

Descriptive metrics also track typical interaction patterns with a given function of a digital product.

More general UX metrics can be divided into the following:

• Quantitative metrics provide insights into user behavior and interaction methods in a quantitative sense.
• Qualitative metrics give insight into reactions, ways of thinking, opinions, emotions, and attitudes (e.g., satisfaction, loyalty, attractiveness, accessibility, credibility).
• Quantitative metrics often measure the behavioral aspect of interaction with a digital product.
• We can use quantitative metrics to measure, among other things:
• Average time required to complete the task
• Percentage of tasks completed successfully
• Number of views and abandonments
• Percentage of user retention
• User error rate
• Number of active users
• Behavioral attitudes.

In turn, we can use qualitative metrics to measure the following:

• Attractiveness
• Credibility
• Trust that a digital product inspires
• Usability (the ease and intuitiveness of use)
• Emotional and cognitive attitudes

What are real user monitoring metrics?

These are metrics or techniques that allow us to track user activity in real-time. They track things like HTTP requests, crashes, or timeouts. Thanks to them, developers can pinpoint exactly what impacts a website's user experience and what to do to improve it.

Why should we use UX metrics?

First, let's start with the fact that UX metrics are a reliable source of insight that can be transformed into useful knowledge.

They're characterized by immediate utility (for use in a specific project and possible future use).

The primary functions of UX metrics provide:

• Evidence of whether something works or doesn't work
• Arguments about how well something works/doesn't work
• Concrete, quantifiable data that allow us to make objective comparisons
• Specifically defined goals and results within a precisely specified time frame
• Material that helps us, through analysis, to observe trends and changes and, therefore, to prepare for them.

Moreover, the use of UX metrics makes it possible to:

• See cause-and-effect relationships by tracking changes — both at the level of the structure of project elements and at the level of the need for continuous optimization of a digital product.
• Diagnose problems, determine their importance, build their hierarchy, and determine the time needed to solve them (e.g., optimize a given function).
• Accurately position a product on the market.
• Make business, design, and technology decisions, both instantaneous and strategic. 

The most important UX metrics

The use of UX metrics, above all, should be rational and driven by established cognitive, research, design, and strategic goals (UX strategy).

It cannot be art, for art's sake, implemented haphazardly.

UX metrics are the most useful when used:

• In a specific time-frame
• For a particular purpose — to support a solution to a specific problem
• In an intended manner, meaning they are driven by the effect we want to achieve.

Examples of key metrics include the following:

• CES (Customer Effort Score)
• CSAT (Customer Satisfaction Score)
• NPS (Net Promoter Score)
• SEQ (Single Ease Question)
• SUPR-Q (Standardized User Experience Percentile Rank Questionnaire)
• SUS (System Usability Scale)
• TPI (Task Performance Indicator)
• Task Success Rate
• Time spent on task
• User Errors

CES (Customer Effort Score)

The Customer Effort Score is a survey used to measure the level of effort required when a user interacts with an application or website's interface.

The user is asked to indicate their feelings on a scale.

In practice, several scales are used; among the most popular are:

• Likert scale
• Numerical scale from 1 to 10
• Rating scale indicating the level of difficulty expressed in words: very difficult, difficult, none, easy, very easy
• Scale with icons in the form of "smiley faces" representing an emotional assessment

CES is a metric addressing Customer Experience that helps understand how easy it is for a customer to interact, achieve a goal, or solve a problem using a product.

CSAT (Customer Satisfaction Score)

The Customer Satisfaction Score is one of the most popular metrics.

It's used to measure the level of satisfaction with using a digital product (or a specific functionality), expressed as a percentage or arithmetic average.

The advantage of CSAT surveys is that they are fast and easy to implement.

NPS (Net Promoter Score)

The Net Promoter Score allows us to estimate satisfaction by indicating how likely a user is to recommend a product to a third party.

Net Promoter Score divides users into three groups:

• Promoters (on a scale they express their desire to promote a site very strongly — 9 or 10 points)
• Passive (7 or 8 points)
• Critics (0 to 6 those who are dissatisfied with the operation of a site)

NPS score represents the difference between the percentage of critics and promoters.

The Net Promoter Score is closely related to user experience — the willingness to make a spontaneous recommendation indicates a positive experience.

SEQ (Single Ease Question)

The Single Ease Question is a survey that allows us (usually on a seven-point scale) to assess the difficulty of performing a given task (such as making a subscription).

SEQ is offered to a user shortly after they complete a task.

The ease/difficulty of performing a single task is strongly related to the time required to complete it.

Users tend to rate tasks as more difficult if they take longer to complete.

SUPR-Q (Standardized User Experience Percentile Rank Questionnaire)

SUPR-Q was created to assess usability, credibility, and user loyalty.

The SUPR-Q questionnaire contains eight questions that address the four most crucial areas from a user experience perspective:

• Usability
• Trust and credibility
• Loyalty
• Attractiveness (of an appearance/aesthetics)

A user provides answers on a Likert scale:

• Strongly disagree
• Disagree
• Neutral
• Agree
• Strongly agree

SUS (System Usability Scale)

SUS is primarily used to measure the usability of digital products and is most often utilized during usability testing.

The System Usability Scale consists of ten questions about usability, the difficulty of use, and learning how the system works.

Respondents answer questions similar to the SUPR-Q questionnaire by using a Likert scale.

TPI (Task Performance Indicator)

The Task Performance Indicator tests crucial tasks a user can perform on a website or mobile app.

Respondents are asked to answer ten questions, which are provided after each task is completed.

The distinguishing feature of this metric is that the survey is repeated after 6 to 12 months to compare results and asks for a determination of the level of confidence in the answer the user gives.

Task Success Rate

The Task Success Rate is an indicator that provides a percentage of the number of tasks a user has completed correctly.

The ability to complete a task correctly and in a satisfactory amount of time affects the user experience rating.

It determines the percentage of users who could complete a task, indicating the intensity of the problem with usability.

The most common tasks that respondents are asked to perform:

• Filling out a form
• Adding a product to a shopping cart in an online store
• Subscribe to a newsletter

Time Spent on Task

The Time Spent on Task is a metric that measures the time required to complete a task.

The less time-consuming the task is, the better the user experience. Users prefer tasks that they can complete quickly and seamlessly.

There are several ways to estimate the time required to complete a task.

It can be:

• The average time of successful completion of a task.
• The average time required for users to give up or perform a task incorrectly.
• The total average time that users spend on a task.

When we know the amount of time required to perform a given action correctly, we have insight into the possible reactions that will occur during that time.

The Time Spent on Task metric measures performance and is used to gain knowledge about the ease of interaction with a digital product.

Measuring the time required to complete a task makes it possible to:

• Diagnose usability problems
• Define the difference between the customer's latency time and the system's latency time
• Define more detailed metrics (e.g., an average time of successful completion or an average time of failure).

User Errors

The User Error indicator allows us to count the number of errors occurring while using a given product's function.

It determines how often a user makes an error while performing a task.

User Errors is a UX metric that determines how easy a feature is to use and how well it performs according to expectations and experiences.

It also helps determine how much users are confused and lost.

Measuring and monitoring the number of errors using the User Errors metric also helps us see and understand user problems and optimize the design based on the data.