barcelona-shartbandi.com
Core Web Vitals & PageSpeed
Measured by Google PageSpeed · metrics on the mobile versionContent & Keywords 38/100
The meta title is the headline shown in search results: one of the most important on-page factors for both rankings and click-through rate (CTR).
Extend the title towards 30–60 characters: a short title wastes precious space for keywords and value proposition.
The meta description is the text shown under the title in search results. It doesn’t directly affect rankings, but it determines how many people click your result.
Add a 70–160 character meta description that summarises the page and invites the click. Each page needs a unique one.
Seeing the page the way searchers see it helps you judge whether the title and description invite the click.
Heading tags give content a hierarchy: they help Google and AI understand what the page is about. The H1 is the main title and should be unique.
Add a unique H1 describing the page’s main content, including the target keyword.
An ordered hierarchy (H1 → H2 → H3, no skips or empty headings) helps Google and AI understand the logical structure of the content, and improves accessibility.
The most frequent words and phrases tell Google what the page is about. They should match the keywords you want to rank for: if you don’t recognise them as relevant, the content isn’t communicating the right topic.
When the text’s most frequent terms also appear in title, description and headings, the message sent to search engines is coherent and strong.
Pages with substantial content tend to rank better: Google and generative AI favour content that answers users’ questions thoroughly.
A nearly empty page rarely ranks: add original, useful text content.
The text-to-HTML ratio shows how much of the page is actual content versus code. Very low values may indicate thin content or bloated markup.
The page has very little text compared to code: enrich the content and slim down the HTML.
The alt attribute describes images to Google (and screen readers). It improves accessibility and lets images rank on Google Images.
Internal links distribute authority across the site and help crawlers discover content. External links to authoritative sources increase trust.
Links with generic text ("click here", "read more") tell Google nothing about the destination page: descriptive anchor text is a relevance signal.
Broken links (404) hurt user experience and waste Google’s crawl budget. They’re among the easiest problems to fix.
Duplicate titles, multiple meta descriptions or a description identical to the title confuse search engines and waste the space available in results.
Keep a single title tag and a single meta description per page, with different, complementary texts.
Technical SEO & Code 95/100
Short, readable URLs with hyphen-separated words are easier for users and search engines to understand, and earn more clicks.
The lang attribute declares the page language: it helps Google serve it to the right audience and screen readers pronounce it correctly.
Hreflang tags tell Google about versions of the page in other languages, preventing duplicate-content issues and serving the right version to each user.
Declaring the charset (usually UTF-8) prevents accented characters and symbols from rendering incorrectly.
The doctype tells the browser how to interpret the page. Without it, the browser falls into "quirks mode" with unpredictable rendering.
Deprecated HTML tags are no longer supported by modern standards and signal dated code, with possible rendering and accessibility issues.
Inline styles (style attribute) bloat the HTML, make code harder to maintain and prevent style reuse via caching.
The favicon appears in browser tabs, bookmarks and Google mobile search results: it strengthens brand recognition.
Add a favicon (48×48 px or multiples) with <link rel="icon" href="/favicon.ico">.
HTML with many syntax errors can be misinterpreted by browsers and crawlers, with unpredictable effects on rendering and indexing.
Tables inside tables are a 1990s-web relic: they slow rendering and make the page harder for crawlers to read.
Flash died in 2020: no modern browser runs it and Google doesn’t index it. Any Flash content is invisible.
The meta keywords tag has been ignored by Google since 2009 and can signal outdated SEO practices. Best removed.
Indexing & Crawling 78/100
The robots.txt file tells crawlers which parts of the site they may crawl. A mistake here can remove the whole site from Google.
Tip: add the line "Sitemap: https://…/sitemap.xml" to robots.txt.
The XML sitemap lists all pages of the site: it helps Google discover and index them faster, especially for large or new sites.
Generate a sitemap.xml (CMSs like WordPress create it automatically), publish it in the root and submit it to Google Search Console.
The canonical tag tells Google which version of the page is the "official" one, avoiding duplicate-content penalties (e.g. URLs with parameters).
Add <link rel="canonical" href="page-URL"> in the <head> of every page.
The noindex directive removes the page from search results. Useful for private pages, disastrous when accidentally left on important ones.
A page-level "nofollow" meta robots stops Google from following any link, cutting the flow of authority to other pages.
The meta refresh (HTML-based redirect) is an obsolete technique: it confuses search engines and degrades user experience.
The site should be reachable from a single version (www or non-www): the other must redirect, or Google sees two duplicate sites.
Set up a 301 redirect from the secondary version to the main domain version.
Every redirect adds a round trip to the server before the page loads, and Google may stop following long chains.
If /page and /page/ both respond 200 without a redirect, Google sees two duplicate pages splitting the ranking between them.
Pick one convention (with or without slash) and 301-redirect the other version.
Non-existent pages must respond with a 404 code. If they return 200 ("soft 404") Google indexes empty pages; a custom 404 page retains visitors.
Configure the server to return a real 404 code for pages that don’t exist.
An AAAA DNS record makes the site reachable over IPv6: better reachability from modern mobile networks and a sign of a well-kept infrastructure.
GEO & AI Search 57/100
The llms.txt file is an emerging standard that helps AI systems (ChatGPT, Claude, Perplexity) understand the site’s structure and content: an advantage for visibility in AI answers.
Create an llms.txt file in the root with a site description and links to key pages in Markdown. It signals care for AI search.
More and more people search through AI (ChatGPT, Perplexity, Google AI). If robots.txt blocks AI crawlers, the site can’t be cited in their answers.
AI crawlers (GPTBot, ClaudeBot, PerplexityBot) don’t execute JavaScript: if the page populates via JS only, it’s an empty page to AI. Google also indexes it with delay.
Pages with FAQPage or HowTo schema are perfect for AI: question and answer already structured, ready to be cited by ChatGPT, Perplexity and Google AI Overviews.
If the page answers frequent questions, add a FAQ section with FAQPage schema: among the most-cited formats by AI.
BreadcrumbList schema shows the site path in search results and helps Google and AI understand the structure and depth of the content.
Add visible breadcrumbs with BreadcrumbList markup: they improve the result snippet and site comprehension.
AI answers questions: a heading like "How much does a website cost?" followed by a clear answer is the format most likely to be cited in AI answers.
Rephrase some H2/H3 as the real questions customers ask, with a direct answer in the first lines below.
Bulleted lists and tables are the formats AI and featured snippets extract most easily: long undifferentiated paragraphs get cited far less.
Turn enumerations and comparisons into bulleted lists or tables: they raise the odds of appearing in featured snippets and AI answers.
Reachable "About" and "Contact" pages are E-E-A-T signals (experience, expertise, authoritativeness, trustworthiness) that Google and AI use to judge how much to trust a site.
Make the About and Contact pages reachable from the footer, with a real address and verifiable references.
Structured data describes content in a format Google and AI understand: it enables rich snippets (stars, FAQs, breadcrumbs) and improves citability in AI answers.
Add JSON-LD markup: at least Organization/LocalBusiness on the homepage and specific types (Article, Product, FAQ…) on other pages.
Backlinks from authoritative sites remain one of the main ranking factors: Open PageRank estimates domain authority (0–10) based on the web’s link graph. Its linking-domain count is partial (it only tallies domains within its own dataset): for full numbers see the “Backlink profile” check.
Build quality backlinks: linkable content (guides, research, tools), digital PR, industry directories and partnerships with authoritative sites.
Total backlinks and above all referring domains (unique domains linking to you) are among the main authority indicators: 100 links from 100 different sites are worth far more than 1,000 links from the same site.
The sameAs property in the Organization schema links the site to its official profiles (LinkedIn, Wikipedia, social): it helps Google and AI recognise the brand as a single trusted entity — a key signal for GEO (AI search).
Add the sameAs property to the Organization schema with the URLs of official profiles (LinkedIn, Instagram, Wikipedia, Crunchbase…).
Publication and update dates (article:published_time or datePublished in the schema) signal freshness to Google and AI: dated or undated content gets cited less.
For editorial content, add article:published_time / article:modified_time or datePublished in the schema.
Speed & Core Web Vitals 86/100
TTFB (Time To First Byte) measures server responsiveness: high values slow everything down.
The HTML download time is the foundation of perceived loading. Beyond 3 seconds, most users abandon the page.
Lightweight HTML downloads and parses faster, especially on mobile.
Compression (GZIP or Brotli) cuts the transfer weight of HTML, CSS and JS by 60–80%%: one of the simplest, most effective optimisations.
Every requested file (images, scripts, stylesheets) is an HTTP request: fewer objects means less waiting and faster loading.
A DOM with too many elements slows rendering and consumes memory, hurting Core Web Vitals.
Minified JavaScript files (no whitespace or comments) weigh 30–60%% less and download faster.
Minified stylesheets reduce transferred weight and speed up first render.
Cache headers tell the browser to reuse already-downloaded files: repeat visits become nearly instant.
Modern formats (WebP, AVIF) weigh 25–50%% less than JPEG/PNG at the same quality: fewer bytes, faster pages.
CSS and JavaScript loaded in a blocking way in the <head> stop the browser from painting the page until downloaded, delaying first render.
HTTP/2 and HTTP/3 download multiple files in parallel over one connection: a free speed upgrade over HTTP/1.1.
Every iframe loads a whole page inside your page: it weighs on loading, and its content isn’t attributed to your site by search engines.
Every external domain you load scripts from adds latency, failure points and privacy risks. Fast sites use very few.
A CDN serves content from the server closest to the user: it cuts latency for distant visitors and absorbs traffic spikes.
Loading fonts from Google servers adds an external connection (slower) and transmits the user’s IP to Google: a GDPR issue already sanctioned in Europe.
The desktop score completes the picture: Google uses mobile for ranking, but desktop experience matters for business users and office workers.
Google’s Lighthouse score summarises the page’s real speed on mobile: it’s the same tool Google uses to assess user experience.
Work through the other checks in this section (LCP, images, render-blocking): the score rises as you fix those.
LCP measures when the page’s main element becomes visible. It’s the most important Core Web Vital: Google uses it as a ranking factor.
Optimise the main above-the-fold element: compress the hero image, preload it with <link rel="preload">, reduce TTFB and blocking CSS.
CLS measures how much the layout "jumps" while loading. Sudden shifts make users click the wrong element and are deeply annoying.
INP (Interaction to Next Paint) measures how quickly the page responds to clicks and taps. Since 2024 it’s an official Core Web Vital, replacing FID.
FCP measures when the first content appears on screen: the user’s first confirmation that the page is coming.
Reduce TTFB and render-blocking resources, enable compression and use font-display: swap for web fonts.
Security 78/100
HTTPS encrypts traffic between user and site. It’s been a ranking factor since 2014 and browsers mark HTTP sites as "not secure".
Anyone typing the address without https:// must be automatically redirected to the secure version, or they browse unencrypted.
The HSTS header forces the browser to always use HTTPS for the domain, protecting against downgrade attacks and cookie hijacking.
Add the header: Strict-Transport-Security: max-age=31536000; includeSubDomains
Resources loaded over http:// on an https:// page are blocked by browsers or trigger security warnings, breaking styles and functionality.
Security headers protect visitors from clickjacking, MIME sniffing and cross-site scripting. They’re also a technical quality signal.
Configure the missing headers: X-Content-Type-Options: nosniff, X-Frame-Options: SAMEORIGIN, Referrer-Policy and a Content-Security-Policy.
Exposing exact server and language versions (e.g. Apache/2.4.41, PHP/7.2) makes it easier to target known vulnerabilities of those versions.
Email addresses exposed in the HTML get harvested by spam bots. Better to obfuscate them or use a contact form.
External links with target="_blank" and no rel="noopener" give the destination page partial access to the origin tab (tabnabbing).
The SPF record declares which servers may send email on behalf of the domain: without it, your emails land in spam and anyone can spoof the sender.
Add a DNS TXT record like: v=spf1 include:<email-provider> ~all
The DMARC record tells providers what to do with emails failing SPF/DKIM checks: it protects the domain from phishing in your name.
Add a TXT record on _dmarc.<domain> like: v=DMARC1; p=quarantine; rua=mailto:dmarc@<domain>
Mobile & UX 100/100
The meta viewport tells mobile browsers how to scale the page. Without it, the site appears miniaturised on smartphones and Google considers it not mobile-friendly (indexing is mobile-first).
CSS media queries adapt the layout to different screen sizes: they’re the foundation of responsive design.
With srcset the browser downloads the right image size for the screen; with loading="lazy" below-the-fold images load only when needed.
The width and height attributes let the browser reserve space before the image arrives: without them, the layout "jumps" (hurting CLS).
Social & Sharing 30/100
Open Graph tags control how the page looks when shared on WhatsApp, Facebook, LinkedIn: a polished preview earns far more clicks.
Add the og:title, og:description, og:image (1200×630 px) and og:url meta tags to control share previews.
Twitter/X Cards define the page preview on X. When absent, X falls back to Open Graph tags if present.
Add <meta name="twitter:card" content="summary_large_image"> alongside the Open Graph tags.
Without an analytics tool you can’t know how many visitors arrive, from where, and what they do: deciding without data is deciding blind.
Install Google Analytics 4 (or a privacy-first alternative like Matomo/Plausible) to measure traffic.
Links to social profiles strengthen brand identity and help Google connect the site with its social entities.
Add links to the brand’s social profiles (typically in the footer).
Compare with a competitor
Analyse another website and see the side-by-side comparison, check by check.
Not happy with the score?
Every issue in this report has a solution. The Rope designs websites that are fast, secure and built to rank — on Google and in AI answers. Let’s talk, no strings attached.