SXO (search experience optimisation) is the work that holds SEO, user experience and conversion together: getting a page found on Google and making sure whoever lands on it completes what they searched for, whether that's a quote request or a purchase. It isn't a Google term and it isn't a ranking factor: it's the name the industry gives to the moment when the people doing SEO and the people designing the website stop working in two separate rooms.
On 10 October 2026 we read the pages Google UK shows on this topic.
Almost all of them say the same thing: SEO and UX are allies, and SXO is the natural next step for SEO.
Half true.
Out of eight pages, not one shows a single case where SEO and UX pull in different directions. And only one links to Google's documentation.
So why do people search for "SEO vs UX"?
Because on real projects the conflict exists, and it almost always shows up in one of these three places:
- the hidden menu the designer wants clean and that, if it's badly coded, Google can't follow;
- the "load more" button on category pages, handy for shoppers and invisible to a crawler that doesn't click;
- the full-screen popup that brings in newsletter sign-ups and that Google lists among the things to avoid.
None of these is solved with a principle. Each one is solved case by case, with Google's written rule in front of you.
We see it on the projects we manage.
Contesini Consulenze, an Italian energy consultancy, had a landing page that was hard to use and no conversion tracking. We rebuilt the contact journey first, then switched the campaigns on: 299 enquiries between September 2023 and March 2024, at €7.20 each, and 68% became a sales opportunity.
Scovaventi, an Italian online shop, had the opposite problem: traffic and zero sales, on a site that was slow and awkward on a phone. After the new Shopify store it brought in €77,000 of sales in a year, with an average basket of €190.
What does that have to do with SXO?
Everything. In both cases the problem wasn't traffic, it was what happened after the click.
In this guide you'll find:
- what SXO is and what Google says about page experience;
- what the UK pages on the topic leave out, with the table from our review;
- the eight points where SEO and UX clash, each with Google's rule;
- who wins when you have to choose;
- the order we apply it in on our projects.
Let's start.
What SXO is (and why Google never mentions it)
SXO is the sum of three jobs that usually have three different owners: SEO brings the person to the page, UX lets them understand where they are and move around, CRO (conversion rate optimisation) takes them through to the action.
The acronym belongs to the industry, not to Google.
The word doesn't appear in any of the Google Search Central documents quoted in this article. What does appear is a sentence that clarifies how experience and ranking relate. On its page experience page, updated on 22 September 2026, Google writes: "There is no single signal".
Put simply: there's no switch called user experience in the ranking. Core Web Vitals are used by the ranking systems, but the same document warns that good results in the reports "doesn't guarantee that your pages will rank at the top".
So UX doesn't count for Google?
It counts, by a different route. Google writes that when there are many helpful results for a search, a great page experience can contribute to success. It weighs in when the competing content is roughly equal, not in place of content.
We measured how much it weighs in practice on the Google UK results for "ux seo": the two fastest pages were seventh and eighth, and the page in first place had the lowest performance score of the group. The full test, Lighthouse on nine results, is in our guide to user experience optimisation; the starting definition is in our page on user experience, and the thresholds in our Core Web Vitals guide.
The point of SXO is something else, and Macropix, an Italian LED wall manufacturer, shows it clearly. The word "ledwall" brought 67,560 impressions and 1,495 organic clicks in twelve months (September 2025 to August 2026) to a category page.
That page doesn't have two separate jobs. It's the same page that has to trigger the quote request.
If SEO pushes it up and UX makes it awkward, the 1,495 clicks stay clicks.
The UK pages on SXO: 0 of 8 show a conflict (and only 1 cites Google)
The method first, then the numbers. On 10 October 2026 we pulled the Google UK mobile results for "sxo" with DataForSEO. Page one shows an AI Overview, a video block and eight organic results, one of which is a LinkedIn post. We read the seven articles plus Abstrakt's guide, a UK agency page cited in the AI Overview, and noted five things for each.
| Page | Words (estimate) | Definition in the first two sentences | Links to Google documents | Concrete case of SEO/UX conflict | Statistics with a source |
|---|---|---|---|---|---|
| digidop.com | 2,000 | No | 7 | No | 0 |
| eminence.ch | 1,400 | Partial | 0 | No | 0 |
| surferseo.com | 2,400 | Partial | 0 | No | 0 |
| precis.com | 300 | Yes | 0 | No | 0 |
| whatissxo.com | 85 | Yes | 0 | No | 0 |
| develtio.com | 1,500 | No | 0 | No | 0 |
| alexisgardin.fr/en | 800 | Yes | 0 | No | 0 |
| weareabstrakt.com | 1,350 | Yes | 0 | No | 0 |
| Total | median 1,375 | 4 of 8 | 1 of 8 | 0 of 8 | 0 of 8 |
The eight pages repeat that SEO and UX get on well.
None says what to do when they don't.
On the Google UK results for "seo vs ux" the picture is half reversed. AWA Digital has a whole section on the conflict, from minimalist navigation to dynamic content, but doesn't link any of it to a primary source. Lyssna, which ranks above it, calls the two "complementary forces" and links no Google documentation.
We took the conflicts those pages list and checked them line by line against Google's documentation.
SEO vs UX: the eight points where they pull in different directions (with Google's rule)
Here's the table we didn't find on page one. Each row is a design choice that makes sense for the people using the site and becomes a risk for the people who need to find it.
| UX choice | What SEO risks | What Google says | How to keep both |
|---|---|---|---|
| Text inside accordions or tabs | That the closed text isn't read | "consider moving content into accordions or tabs to save space" | Accordions yes, with the text already in the HTML on load, not loaded on click |
| "Load more" or infinite scroll on category pages | Products beyond the first batch never discovered | "Google's crawlers don't "click" buttons" | The button stays, but each batch has its own URL (?page=2) linked with an HTML link |
| Hidden menu or menu built in JavaScript | Links Google doesn't extract | "Google can only crawl your link if it's an <a> HTML element" | The menu can stay hidden if the code has real links, plus links in the text |
| Combinable filters (size, colour, price) | Thousands of useless URLs to crawl | Crawlers access "a very large number of faceted navigation URLs" | Index only the combinations people search for; the rest after the # or blocked in robots.txt |
| Full-screen popup | A page that's hard to understand, for the search engine too | "Don't obscure the entire page with interstitials" | A banner on a small part of the screen; mandatory gates, such as age verification, are exempt |
| Automatic redirect by language or IP | Googlebot sees only one version of the site | "Avoid automatically redirecting users" | A banner suggesting the language, and hreflang between the versions |
| Short, concise page | The fear of having "too little text" | On a preferred word count: "(No, we don't.)" | The questions of the people searching decide the length |
| Copy written "for the keyword" | Readability and spam risk | "Repeating the same words or phrases so often that it sounds unnatural" | Write for the reader: keyword in the title, the H1 and wherever it's really needed |
The most misunderstood case is the first. It's Google's own document on mobile-first indexing that suggests accordions and tabs to save space on a phone. The same document sets the limit: "Google won't load content that requires user interactions". Text closed inside an accordion counts. Text that only arrives after the click doesn't. The rest of the rules are in our mobile-first indexing guide.
The second is the one that costs online shops most. Google's document on pagination says it plainly: crawlers don't click buttons and generally don't trigger JavaScript functions. On a category with a "load more" that has no URL, the products Google is likely to see are only those in the first batch.
And how much do category pages matter?
A lot. In our test on seven commercial UK results pages, 67% of page one was made up of category pages: 41 results out of 61, against 1 product page.
Menus and filters are the same problem seen from two sides. Google reliably extracts only links written as <a> elements with an href: a menu that opens pages with a JavaScript event can look beautiful and, to the crawler, be empty. Filters do the opposite and generate too many URLs, and Google's guide to faceted navigation explains that time spent on the useless ones is time taken away from new pages. How to distribute the links that matter is in our internal linking guide.
Popups and language redirects are the two cases where the rule is clear-cut. Google writes that intrusive interstitials "may lead to poor search performance", and asks you not to redirect automatically based on the language the user is assumed to speak. The answer for multilingual sites goes through hreflang.
The last two rows settle an old misunderstanding. Asked whether Google prefers a certain length, the guide to helpful content replies "(No, we don't.)". And the spam policies define keyword stuffing as words repeated "so often that it sounds unnatural". We've measured length and density in our guides to blog post length and keyword stuffing.
How many products in your catalogue can't Google see?
You've just seen that a "load more" with no URL hides everything below the first batch from Google, and that a JavaScript menu may lead nowhere. We check your site against the eight points in the table and tell you which ones are costing you visits and which are just a matter of taste.
Book a strategy consultationWho wins when you have to choose (almost always neither)
Read the eight rows again and the verdict is simpler than it looks.
In five cases out of eight there's nothing to choose. Accordions, "load more", menus, filters and language are solved in the code, without taking anything away from visitors.
In the other three UX wins, and Google agrees. Popups, text length and repeated keywords: what Google writes for search matches what readers would prefer.
So the conflict doesn't exist?
It exists, but it's a conflict of timing, not of goals. SEO comes in when the site has already been signed off, and at that point every fix is a cost.
At Scovaventi the previous shop had an unclear structure, a CMS that slowed it down and no attention to mobile. Problems built in at the design stage, not later. They were only solved by rebuilding the shop.
At Bosatta, an Italian online shop, we followed users from product search to payment to see where they stopped, and made the purchase journey clearer on a phone. Together with separate Google Ads campaigns for consumers and workshops, twelve months brought €240,000 of sales and 127 enquiries from resellers, across four markets.
In the UK this order matters too, and there are two numbers that explain why. According to Ofcom's Adults' Media Use and Attitudes 2026 report (7,533 adults aged 16+, fieldwork September to November 2025, published 2 April 2026), one online adult in five (20%) goes online only through a smartphone. And the Lloyds Essential Digital Skills 2025 survey, carried out by Ipsos on 4,175 adults, finds that 85% of UK adults have foundation-level digital skills: roughly one in seven doesn't.
With an audience like that, every extra step between the search and the form loses people that a test run on confident desktop users wouldn't count. If you have legal obligations, the rules are in our guide to web accessibility.
From visit to enquiry: the part SEO doesn't measure (and where CRO comes in)
SEO stops at the click. SXO doesn't.
On the purchase side the most used reference is Baymard Institute: the documented average cart abandonment rate is 70.22%, calculated across 50 studies. In its survey of US shoppers, 17% abandoned an order because the checkout was too long or complicated and 18% because the site wanted them to create an account.
Baymard also estimates that a large online shop could gain 35.26% in conversion rate with a better-designed checkout, based on tests on large US and European sites. These are figures from markets other than the UK: we haven't found an equivalent public UK figure built with the same method.
For lead generation the mechanism is the same, only shorter. At Contesini the order was: journey analysis, a new landing page that's simpler on a phone, tracking, and only then the campaigns. The return turned positive in the first few weeks.
And if the tracking is wrong?
Then CRO optimises the wrong number. At Valore Solare, an Italian solar installer, when we took over the account it showed 15,267 conversions against 5,467 clicks in one month: they were technical events on the site counted as enquiries. Before any test on the page, the conversion has to be redefined. How to calculate the conversion rate, and how to tell whether a website isn't converting or just has little traffic, we've covered in two separate guides.
How to do SXO in practice (the order we use on our projects)
Five steps, in this order. The third is the one almost everyone skips.
- Start from the queries, not the design. In Search Console you export the searches that bring impressions to the page and check whether the page answers that question. The method for reading them is in our guide to search intent.
- One page, one question, one action. At Valore Solare every audience got the same ads and the same landing page; today the campaigns are split by intent, and of 1,506 enquiries in eleven months 1,188 came through the landing page form.
- Run through the eight points in the table before launch, not after. They cost an hour at prototype stage and a rebuild once the site is live.
- Track a single conversion as the primary one. At Macropix the only optimisation conversion is the completed enquiry form. Together with campaigns reorganised by product family, the cost per enquiry fell from €94.37 in 2024 to €39.44 in 2025.
- Measure on the queries coming in, not on scores. At Sunscape, in the six months after the region-based structure went live, keywords on page one went from 1 to 38 out of 143, impressions from 329 to 5,809 a month, clicks from 46 to 195. The timing of each phase is in our guide to how long SEO takes.
Summary (and one thing the industry rarely says)
SXO is SEO plus UX plus conversion, with a single owner.
It isn't a Google signal: "There is no single signal".
Zero UK pages out of eight show a real conflict.
Five conflicts out of eight are solved in the code.
UX wins the other three, with Google's agreement.
The thing that's rarely said: SXO doesn't need a new job title, it needs a meeting moved earlier. In the eight cases in the table the fix costs little if it's decided while the menu and the category grid are being designed, and costs a new shop if it's discovered after launch, as happened to Scovaventi. The acronym is only useful if it changes who's at the table when the prototype is signed off.
If you're about to rebuild your website or a section of your catalogue, that's where our SEO work starts. If the traffic is there and the enquiries aren't, we start from the Google Ads campaigns and the landing page, as we did for Contesini. See you next time!
Frequently asked questions
It depends on how you work. By definition SXO includes SEO and adds page experience and conversion. In practice the difference is where you stop measuring: SEO looks at impressions and clicks, SXO goes as far as the enquiry or the order. Google doesn't use the term and there is no ranking signal called SXO.
No, not as a single signal. Google's documentation on page experience says there is no single signal: Core Web Vitals are used by the ranking systems, but good scores don't guarantee top positions. Experience counts most when many pages answer the same search well.
Yes, if it's already in the page code when the page loads. Google itself suggests accordions and tabs to save space on mobile, but it won't load content that requires an interaction such as a click or a swipe. Text that is closed but present counts; text that only arrives after the click doesn't.
It depends on how it's built. Google's crawlers don't click buttons and generally don't trigger JavaScript functions, so products loaded by scrolling or by a load more button stay invisible unless they have their own URL. The fix is to give each batch an address, for example ?page=2, linked with a normal HTML link.
It depends on the popup. Google asks you not to cover the whole page with interstitials and says they may lead to poor search performance. A banner that takes up a small part of the screen is fine, and mandatory gates such as age verification are exempt.
No. Five of the eight conflicts we checked are solved in the code without touching the design: HTML links in the menu, a URL for each paginated page, text present in the HTML, managed filters and no automatic language redirects. A rebuild is needed when the problem is in the structure, as with an online shop that is slow and awkward on mobile.
