Google PageSpeed: what 80+ means for your business

Most business owners see PageSpeed for the first time when someone sends them a red 34 and tells them "your site is slow". The number is scary, but it's rarely explained. Below you'll see exactly what it measures, what it means for your sales, and when it's worth worrying about.

What exactly does PageSpeed measure?

PageSpeed Insights runs an automated visit to your page and produces four scores: Performance (speed), Accessibility, Best Practices, and SEO. When we say "the site's PageSpeed", we almost always mean Performance, because that directly affects the visitor.

Put simply, the four scores answer four questions:

• Performance: How quickly does the visitor see and get to use the page?
• Accessibility: Can someone using a screen reader or with low vision use it?
• Best Practices: Is it built in a safe, modern way (HTTPS, no browser errors)?
• SEO: Can Google read and understand it properly?

An interesting development from 2026 is the experimental Agentic Browsing category in Lighthouse, the tool PageSpeed Insights also uses. It checks elements that make it easier for an AI agent to interact with a website, such as proper button labelling, layout stability, and the ability to use machine-readable tools. It doesn't have the classic 0–100 score but shows checks that pass or fail. It isn't a fifth established PageSpeed Insights score. But it's a sign of where the web is heading: we design for people, search engines, and increasingly for digital assistants. Source: Chrome for Developers.

The metrics, in business language

The Performance score comes from five measurements. The four worth knowing:

• FCP (First Contentful Paint), "when I see something": the moment the first text or image appears, instead of a blank screen.
• LCP (Largest Contentful Paint), "when I see the main content": when the largest element appears, usually the main photo or the title.
• TBT (Total Blocking Time), "how much the page blocks": how long the browser's main thread stays busy with heavy work and may delay responsiveness.
• CLS (Cumulative Layout Shift), "whether the page jumps around": whether elements move while it loads, and you tap the wrong button by mistake.

TBT carries the most weight (30%), followed by LCP and CLS (25% each), while FCP and Speed Index contribute 10% each. INP (Interaction to Next Paint) doesn't go directly into the lab score: it's a Core Web Vital that shows how fast a page responds to real users' actions. Source: Lighthouse.

When is it green, orange or red

• FCP: green up to 1.8 s · orange 1.8–3 s · red over 3 s
• LCP: green up to 2.5 s · orange 2.5–4 s · red over 4 s
• TBT: green up to 200 ms · orange 200–600 ms · red over 600 ms
• CLS: green up to 0.1 · orange 0.1–0.25 · red over 0.25
• INP (real users): green up to 200 ms · orange over 200 to 500 ms · red over 500 ms
• Overall Performance score: green 90–100 · orange 50–89 · red 0–49

Why is the mobile score lower than desktop?

Because the mobile lab test simulates a device and connection with limited resources, while desktop is tested under more favourable conditions. That's why the same page can show, for example, 99 on desktop and 60 on mobile.

That's no reason to ignore desktop: you need good performance on both versions. But mobile is the more demanding test, and that's where we've chosen to give our guarantee. If someone shows you only the desktop score, you don't have the full picture.

Does PageSpeed affect your position on Google?

The essence is this: the tool's Performance score isn't, on its own, a ranking factor. Core Web Vitals from real users are used by Google's ranking systems, but they aren't enough for high positions: content relevance and quality remain decisive. That's exactly what Google Search Central states.

To understand it, you need to separate two kinds of data PageSpeed shows:

• Lab data: a simulated visit at the moment you press "Analyse". This is where the 0–100 score comes from. It's ideal for diagnosis, but it isn't real people.
• Field (real-user) data: aggregated measurements from Chrome users over a rolling 28-day window, via the Chrome UX Report. Here you look at LCP, INP (responsiveness to actions) and CLS. Core Web Vitals are part of the page experience signals Google uses.

What does "No Data" mean? That there isn't enough suitable data for public reporting in the Chrome UX Report — often because the page is new or has little traffic. The tool may show data for the whole site instead of the specific page. If even that isn't enough, you rely temporarily on the lab test for technical diagnosis, without confusing it with the real experience of all visitors.

The conclusion: speed doesn't replace good content. But when two businesses offer equally relevant answers, the better page experience can be an added advantage.

What are the real benefits of a fast site?

The biggest gain isn't a green circle in the tool. It's that the visitor reaches the content quickly, can get in touch, and completes a purchase or booking more easily. A fast, well-structured website cuts needless delays, helps campaigns, and makes it easier for machines to access the content. That's performance for the business, not just a technical number.

Fewer visitors leave before the page loads

A page that lags increases the risk of the visitor leaving before seeing your offer. Think of someone tapping a paid ad from their phone and waiting, unable to read or tap anything: the campaign has already paid for the click, but the lead may be lost before it starts. Buying traffic isn't enough; the site has to be able to make use of it.

More calls, forms and purchases

In the study "Milliseconds Make Millions", carried out by Deloitte and fifty-five on behalf of Google across 37 businesses, an improvement of 0.1 seconds across four speed metrics on mobile was linked to a rise in conversions of 8.4% in retail and 10.1% in travel. These are findings from a specific study, not a promise that every site will achieve the same percentages. But the business point is strong: even small delays are worth measuring. And that holds whether you're after purchases or phone calls and completed forms.

Better ad performance

In Google Ads, the landing page experience is one of the three components of Quality Score. Google clarifies that Quality Score is a diagnostic indicator and not a direct parameter of the auction. Even so, a page that loads fast and answers clearly what the ad promises can offer a better experience and more completed actions with the same ad budget. We don't promise a lower cost per click with a number; we improve where the click lands.

Better reading by AI and crawlers

This is where speed connects with AI Visibility. Many crawlers that collect content for tools like ChatGPT, Perplexity and Claude don't execute JavaScript, as Vercel research in 2024 showed. If the substantive information appears only after scripts run, part of it may not be accessible to them. That's why at Praion we put weight on the website's architecture: with Next.js, proper rendering of content in the initial HTML, sound structure and clean code, so that people, Google and AI crawlers are all served. Easier access doesn't guarantee an AI will cite your site, but it's the right technical foundation for it to be able to use it.

What does a score of 80, 90 or 98 mean in practice?

80+ is the minimum measurable threshold we take on contractually at Praion, not Google's official "good" mark. Google colours 90–100 green. 98, on the other hand, is an excellent result on that particular test, without meaning every visitor will have exactly the same experience. Our goal is for speed to support the real operation of the website, not to chase a number for no reason.

• Below 50: Lab performance is low. There's a serious reason to check the delays and possible loss of visitors.
• 50–79: There's room for improvement. Often a few interventions on images, scripts or loading have a substantial effect.
• 80–89: It stays orange in the tool, but it's above the minimum threshold of the written Praion 80+ guarantee. In practice we aim for higher scores.
• 90–100: Google's official green category. Particularly high scores are a strong indication of good lab performance, but new videos, pixels or tools can always affect it.

Note something: one measurement isn't final. The score can vary from test to test due to load, network and script execution. That's why we've set a transparent process for our guarantee: five consecutive measurements per page, with the median as the criterion — the middle of the five, not the best result.

From contract to practice: results of 96–98

80+ is the minimum we guarantee, not the goal we settle for. Speed is designed from the start — from how we organise the code to when each image loads. Here are three published mobile Performance measurements, with the date they were recorded:

winnert.gr: 98 on mobile (26 Aug 2026). Proof that the 80+ guarantee doesn't limit us to the minimum.

praion.gr: 96 on mobile (11 Sep 2026). Our own website, with the same development logic we apply for our clients.

ygeiapaidi.gr: 97 on mobile (23 Sep 2026). Another project that significantly exceeds the contractual minimum.

What we do differently:

1. We optimise every image into modern formats, at the right size for each screen, and load the one shown at the top first.
2. We limit unnecessary scripts. Whatever's needed loads with the right priority, so core information and user actions aren't delayed needlessly.
3. We build with Next.js and our own CMS, avoiding needless plugins and making sure the substantive content is rendered properly in the HTML.
4. We choose suitable hosting and infrastructure, reducing response time as far as possible for the markets the site serves.
5. We lock the dimensions of images and elements, so the page doesn't "jump" while it loads.

Why do most sites have a low score?

Many sites get heavier over time: more images, features, plugins and third-party tools. The following are common causes worth checking:

• Heavy off-the-shelf themes that load dozens of features "just in case", even if you use three.
• Page builders (drag-and-drop type) that produce far more code than the page needs.
• Dozens of plugins, each with its own files, loading on every page.
• Uncompressed photos straight from the phone or camera, 4–5 MB each.
• Third-party scripts: chat widgets, Facebook and TikTok pixels, maps, cookie banners, all loading before the content.

How to check your own site (in 1 minute)

You can do a first check at pagespeed.web.dev without an account and without technical knowledge. For a conclusion, though, look at the individual metrics too, not just the big number.

1. Go to pagespeed.web.dev.
2. Paste the URL of your homepage or the page that brings the most visits and press "Analyse".
3. Select "Mobile": look at Performance and LCP, TBT, CLS, FCP. If you repeat the test, compare the results; for our guarantee we apply five measurements and keep the median value.

If you see below 50, the page needs priority in the technical check. In the 50–79 range, there are often specific high-value interventions. It doesn't mean you necessarily need a new website: often we improve the existing implementation; other times a more substantial change in architecture is required.

Praion's 80+ guarantee

Our difference isn't that we simply say "we build fast sites". For every new Praion website, the minimum 80+ Performance on mobile is written into the contract. We check the homepage and one page from each core content type the project includes, with five measurements per page and the median value. If the threshold isn't met, we fix it at no extra charge before delivery closes. The commitment covers the final site at delivery, under the specific guarantee terms. 80+ is the floor, not the ceiling: our published measurements show 96–98. If you already have a website, request a free measurement to find out what's really worth improving.