User experience: what it is, what it is made of and how to measure it

User experience is a person's perceptions and responses that result from the use, or anticipated use, of a product, system or service.

The definition comes from the ISO 9241-210:2019 standard, "Ergonomics of human-system interaction, Part 210: Human-centred design for interactive systems", reviewed and confirmed in 2025. The practical consequence is concrete: you do not assess user experience by looking at your website, you assess it by watching someone try to get something done on it. Only part of what that person sees ends up in Google's ranking systems, and the table further down shows which part.

What it is made of

The standard separates three variables, and it pays to keep them separate when you work on a site too.

• The person: what they expect before they arrive, how well they know the sector, which habits they bring from other sites. Someone who has bought industrial spare parts for twenty years looks for the part number, not the photo.
• The system: interface, content, response times, information architecture, the tone of error messages.
• The context of use: where that person is while using the site. A quote request form filled in at the office on two screens is not the same form filled in on a building site with one hand and two bars of signal.

There is also a dimension that almost always disappears from website discussions: the experience includes what happens before and after. The expectation set by a Google Ads ad is part of the landing page experience. The order confirmation email received three days later is part of the e-commerce experience. If traffic arrives on the back of a promise the page does not keep, you have an experience problem even with a technically perfect site, and it is the same problem SEO calls a mismatch with search intent.

Usability and interface are smaller things

The three terms get used as synonyms, and they are not. The most quoted definition in the industry is the one Don Norman and Jakob Nielsen put in writing in August 1998 for the Nielsen Norman Group: user experience covers all aspects of a person's interaction with the company, its services and its products. In the same article they give an example that still holds: a film review site can have a flawless interface and still be a poor experience if its database only lists releases from the big studios.

Usability is a quality attribute of the interface, so a subset. Nielsen breaks it down into five components: learnability, efficiency, memorability, errors and satisfaction. The same text is the source of the figure many people quote and few apply: five test users are enough to find the most serious problems in an interface.

The user interface is the layer below that: buttons, colours, grid, states. You can rebuild a UI from scratch and leave a broken user experience untouched, because the problem was an incomplete catalogue or delivery times nobody stated.

How much of this Google sees

Far less than you will read elsewhere. Google Search Central's page on page experience, in the version updated in December 2025, is clear: there is no single page experience signal. It lists six aspects to check, and only the first is used directly by the ranking systems.

• Good Core Web Vitals (the only one used directly in ranking)
• pages served over HTTPS
• displays well on mobile devices
• no excessive ads that get in the way of the main content
• no intrusive interstitials
• a clear distinction between the main content and the rest of the page

There are three Core Web Vitals, with public thresholds documented on web.dev: LCP within 2.5 seconds, INP within 200 milliseconds, CLS within 0.1. Each threshold is assessed at the 75th percentile of page loads, separately for mobile and desktop, and all three have to be met. Since 2024, INP has replaced FID as the interactivity metric, something half the articles still in circulation have not caught up with. We have covered thresholds, typical causes and the fixes that pay off in the entry on page speed. The point here is different: these three numbers cover a thin slice of what a person experiences when using a site.

Usability, clear language, a complete catalogue, easy returns: none of these is a declared signal. They reach search engines indirectly, through the links a site earns, branded searches and conversions. In our list of ranking factors, page experience sits among the supporting signals, not the decisive ones.

How to measure it, piece by piece

There is no single number. Each component has its own tool, and a different person in the company usually looks after it.

UX component	What you look at	Which tool	Does Google use it as a signal?
Perceived speed and responsiveness	LCP, INP, CLS on real traffic data	Chrome UX Report, PageSpeed Insights, Search Console	Yes, directly
Mobile display	Readability, tap targets, cut-off content	Testing on a real device, not just an emulator	Yes, one of the six page experience aspects
Connection security	HTTPS on every resource, mixed content	Certificate, browser console	Yes, one of the six aspects
Page clutter	Interstitials, ad density, the boundary of the main content	Manual inspection on mobile	Yes, one of the six aspects
Usability	Learnability, efficiency, memorability, errors, satisfaction	Tests with five users on real tasks, completion rate	No
Accessibility	WCAG 2.2 conformance, keyboard use, screen readers	Automated audit plus manual checks	No
Stated satisfaction	The person's verdict after the task	SUS questionnaire, CSAT	No

SUS deserves a couple of lines, because it is the cheapest tool on the list and one you rarely see used on business websites. John Brooke wrote the System Usability Scale in 1986: ten statements, five levels of agreement, a final score from 0 to 100 that is not a percentage. Jeff Sauro worked out its benchmark across roughly 500 studies and more than 5,000 people: the average is 68. Below 68, your form is harder work than average, and you can find that out by asking eight people to fill in ten lines.

Three numbers that do not measure user experience

This is the part that is almost always missing, and it leads to the worst decisions.

• Bounce rate. A visit with a single event can mean the page was useless or that it answered the question in four seconds. On a glossary entry like this one, a high bounce rate is the expected behaviour.
• Average time on page. It goes up both when the copy is interesting and when the price list is unreadable. On its own it cannot tell reading from struggling.
• The PageSpeed Insights score. It is lab data, produced by Lighthouse from a single simulated run. The Core Web Vitals that count for ranking come from the Chrome UX Report, that is, from real traffic from eligible Chrome users. Two sections of the same screen, two different things.

The numbers that do tell you something are duller: task completion rate, errors per attempt, how many people use site search and what they look for, how many support tickets come from the same screen. If you already track them on a dashboard, they belong next to your other SEO KPIs, not in a separate spreadsheet.

In the UK, accessibility is not a bonus

Web accessibility is the part of user experience that has a legal reference point, and for some organisations it is an obligation. The technical benchmark is the Web Content Accessibility Guidelines 2.2, a W3C Recommendation in the version dated 12 December 2024, organised into three conformance levels: A, AA and AAA.

In the UK, the Public Sector Bodies (Websites and Mobile Applications) (No. 2) Accessibility Regulations 2018 require public sector websites and apps to meet WCAG 2.2 AA and to publish an accessibility statement. The Government Digital Service monitors a sample of sites each year, and enforcement sits with the EHRC in Great Britain and the ECNI in Northern Ireland (GOV.UK guidance). Private businesses are not covered by those regulations, but the Equality Act 2010 already requires every service provider to make reasonable adjustments for disabled people, and UK companies selling to consumers in the EU also fall under the European Accessibility Act. The practical effect is the same for everyone: a 2:1 contrast ratio on a buy button makes life harder for all users, not only for people with a registered disability.

Where user experience meets SEO work

Three real points of contact.

• Site structure. The choices that make a catalogue easy for a person to navigate make it easy for a crawler to crawl, and breadcrumbs are the clearest case: the same line serves the visitor and the crawler. This is the common ground between technical SEO and design, and it is decided when you build the site, not afterwards.
• Matching the query. A page that ranks for a commercial query and opens with a thousand words of company history has an experience problem before it has an optimisation problem.
• Performance measured in the field. The only part of UX that feeds into ranking systems, and the only one you can check without talking to anyone.

The concrete fixes, in order of return, are in the guide to user experience optimisation. If you would rather start by finding out where you lose traffic and conversions before touching the design, begin with an SEO audit of the site.

Something the industry rarely says: the first signs of a broken user experience do not come from SEO tools, they come from site search queries and the support inbox. If thirty people a month search for "returns" in your online shop's search box, the link to the returns page is in the wrong place, and no Core Web Vitals report will tell you. It works the other way too, which is why green dashboards mislead: field data is read at the 75th percentile, so one person in four is always having a worse time than the number you are looking at. That is the person who emails support.

Frequently asked questions