Market OS glossary
Glossary: in plain words.
Every term that appears in our tools, reports and guides, explained without jargon. Plus the exact formula we use to score your page.
Updated on 2026-11-05
How we score
- Score (0 to 100)
- Summarizes how well the page does on everything we measure. We first score each section, then take the weighted average of the sections with data (heavier ones, like GEO / AI or technical SEO, count more). What we couldn't measure neither adds nor subtracts.
- Score formula
- Each check is worth pass = 1, improvable = 0.5 and missing = 0, multiplied by its weight. Section score = Σ(weight × value) ÷ Σ(weight), on a 0–100 scale. Overall score = weighted average of the sections with data. Section weights: GEO / AI 20, technical SEO 18, Speed 12, Passive security 12, Content and intent 12, Structured data 8, Conversion 8, Measurement 7, Basic accessibility 3.
- Points (+pts)
- How much your overall score would go up if you fixed only that check and everything else stayed the same. We compute it by redoing the math with that check set to pass. It's how we sort: first what raises the score most for the least effort.
- Grades A–F
- The same score as a letter so it reads fast: A from 90 to 100 (excellent), B 80–89 (good), C 65–79 (fair), D 50–64 (weak) and F under 50 (urgent).
- Pass, improvable, missing, not measured
- Pass: it's right. Improvable: it's there but could be better (worth half). Missing: absent or wrong. Not measured: we couldn't verify it from outside (for example, the page blocked the measurement); it doesn't count in the score.
- Effort
- Approximate time to fix a check: low (hours), medium (days) or high (weeks). It depends on your site and who maintains it; it's a guide for prioritizing, not a quote.
Speed
- Lighthouse and PageSpeed Insights
- Lighthouse is Google's tool that measures a page's speed in a lab (a simulated phone). PageSpeed Insights runs it for you and also shows real-user data when Google has it. Its speed score goes from 0 to 100; 90 or more is good.
- Core Web Vitals
- The three experience metrics Google uses for ranking: LCP (how fast the main content shows), INP (how fast it responds to taps) and CLS (how much the page jumps while loading).
- LCP (Largest Contentful Paint)
- Time until the largest element on screen shows, usually the main image or headline. Good: 2.5 seconds or less.
- INP (Interaction to Next Paint)
- How long the page takes to react when you tap a button or type. Good: 200 milliseconds or less. Only measured with real users.
- CLS (Cumulative Layout Shift)
- How much content moves while loading (the button that shifts right when you're about to tap it). Good: 0.1 or less.
- TBT (Total Blocking Time)
- Time JavaScript keeps the page busy so it can't respond. It's the lab stand-in for INP. Good: 200 milliseconds or less.
- TTFB (Time to First Byte)
- How long the server takes to start answering. If it's slow, everything else is delayed. Good: under 800 milliseconds.
- CrUX (real-user data)
- Chrome User Experience Report: speed measurements from real people visiting your site with Chrome. Google uses them for Core Web Vitals; low-traffic sites don't have enough data.
- Page weight
- What gets downloaded to show it: HTML, images, fonts and scripts. Every extra MB costs seconds on a mobile connection. Formats like WebP and AVIF and compression (gzip or brotli) reduce it.
- Minification (JavaScript and CSS)
- Removing the spaces, line breaks and comments that only help whoever writes the code. The file does the same but weighs less, so the page loads sooner. It runs automatically when publishing, with tools like esbuild or terser; the code you edit stays readable.
Technical SEO
- SEO
- Search engine optimization: making sure Google and Bing can find, understand and show your pages for your customers' searches.
- Page title
- The blue text Google shows in its results and the tab name. It should say what you offer and stay under about 60 characters so it isn't cut.
- Meta description
- The paragraph Google may show under the title. It doesn't rank by itself, but it decides whether people click. Ideal: 120 to 155 characters.
- H1 and headings
- The H1 is the main heading on the page; there should be one. H2 and H3 are subheadings that organize content for people, Google and AI assistants.
- Canonical
- A tag telling Google which address is the official one when copies exist (with and without www, with parameters). It keeps signals from splitting across several URLs.
- noindex
- An instruction asking search engines not to show the page in their results. Useful for internal pages; a mistake on pages you want to rank.
- robots.txt
- A file at the root of the site (yoursite.com/robots.txt) telling Google, Bing and AI crawlers what they may visit. A mistake there can remove the whole site from search.
- Sitemap (sitemap.xml)
- A list of the pages you want search engines to know about, with their update date. It helps them find new pages faster.
- 301 redirect and status codes
- The server answers with a number: 200 means «here it is», 301 «moved permanently to another address», 404 «doesn't exist» and 5xx a server error. Several redirects in a row (a chain) slow everything down.
- Internal links and depth
- Internal links go from one page of your site to another. They are how people and search engines find your pages and learn which ones matter. Depth is how many clicks a page is from the home page: 3 or fewer is ideal.
- hreflang and language
- Tags that connect a page's versions in different languages (for example, Spanish and English) so Google shows the right one to each person.
- Open Graph
- Tags defining the title, description and image when someone shares your page on WhatsApp, LinkedIn or Facebook. Without them, the card is empty or shows a random image.
- Alt text
- A short description of an image for screen readers and search engines. If the image doesn't load, it's also what gets shown.
- Doorway pages
- Near-identical pages made in series to rank for each city or search variant, swapping only the place name. Google treats them as spam and may demote the whole group. One page per city is fine if it has content specific to that place.
GEO and AI search
- GEO (Generative Engine Optimization)
- Optimizing your content so AI assistants like ChatGPT, Gemini, Perplexity or Google's AI Overviews find it, understand it and cite it in their answers.
- AI Overviews
- The AI-generated summaries Google shows above some results, with links to the sources it used.
- AI crawlers
- Programs from OpenAI, Anthropic, Google, Perplexity and others that visit pages to answer questions (OAI-SearchBot, Claude-User, PerplexityBot) or to train models (GPTBot, ClaudeBot, Google-Extended). If you block them, they can't cite you.
- AI mention and citation rate
- The mention rate is the share of ChatGPT, Gemini, Claude or Perplexity answers that name your brand; the citation rate, the ones that link a page on your site. Since AI doesn't always answer the same way, each question is asked several times and counted as «x of N».
- Citation gap
- Sites AI assistants cite when they recommend a competitor and not you: reviews, directories, articles. Getting listed there is one of the things that helps AI name you most.
- llms.txt
- A Markdown file at the root of the site with a summary and a list of your important pages, meant for AI assistants. It's an open proposal: optional and cheap, no magic promises.
- Markdown version and JavaScript content
- Many AI crawlers don't run JavaScript: if your text only appears after it runs, they don't see it. Serving the content in the HTML (or a Markdown version) makes it readable for them.
- Citability
- How easy it is for an AI assistant to take a passage from your page and use it as an answer: sections that open with the answer, stand on their own, include sourced data and aren't too long.
- E-E-A-T
- Experience, expertise, authoritativeness and trust: what Google and assistants look for before trusting content. It shows through a real author, date, sources, cases and contact details.
- Entity, sameAs and Wikidata
- To search engines your brand is an «entity»: something with a name, site, profiles and verifiable data. The sameAs property links your official profiles, and Wikidata is a public database Google and assistants use to recognize you.
- Readability (Fernández-Huerta)
- How easy a Spanish text is to read, based on sentence and word length. The Fernández-Huerta index is the Spanish version of Flesch: 60 or more is easy, 50 to 60 fairly difficult and under 50 difficult, like an academic text.
Structured data
- Structured data (schema.org, JSON-LD)
- Invisible code that describes your page in a format machines understand: who you are (Organization), what service you offer, FAQs, breadcrumbs. JSON-LD is Google's recommended way to write it.
- Frequently asked questions (FAQPage)
- A visible question-and-answer section on the page, which can be marked up with FAQPage structured data. It's among what AI assistants and Bing reuse most. Google retired the FAQ rich result on May 7, 2026, so FAQPage no longer changes how you appear on Google: it's optional and costs no points in Market.
Security and email
- HTTPS and TLS certificate
- HTTPS encrypts the connection between the visitor and your site; a valid TLS certificate makes it possible. Without it, browsers mark the site «Not secure».
- HSTS
- A header telling the browser «always open this site over HTTPS», even if someone types http://. It prevents attacks that intercept the first visit.
- Security headers (CSP and others)
- Instructions the server sends with each page so the browser protects it: CSP limits where scripts load from, X-Frame-Options stops others framing your site, nosniff and Referrer-Policy close minor leaks.
- Mixed content
- An HTTPS page that loads images or scripts over http://. Browsers block them or show warnings.
- Subresource Integrity (SRI)
- An integrity attribute on a script or link tag with the fingerprint (hash) of the expected file. If a library you load from a CDN changes without your consent, the browser won't run it. It only applies to pinned-version files; Google Tag Manager and Analytics don't support it.
- SPF
- A DNS record listing which servers may send email with your domain. Without it, impersonating you is easier and your email lands in spam.
- DKIM
- A digital signature your server adds to each email; the recipient checks it with a key published in your DNS. It proves the message wasn't altered.
- DMARC
- A DNS policy saying what to do with email that fails SPF and DKIM (nothing, spam or reject) and where to send reports. Gmail and Yahoo require it from bulk senders.
Measurement
- Google Search Console
- Google's free tool showing which searches you appear for, how many clicks you get, which pages are indexed and what errors it found.
- Bing Webmaster Tools
- Bing's equivalent of Search Console. It matters because Bing powers ChatGPT Search and Copilot.
- IndexNow
- An instant ping to Bing, Yandex, Seznam and Naver when you publish or change a page. You enable it by publishing a key at yoursite.com/<key>.txt. Google doesn't use it.
- GA4 and Google Tag Manager
- Google Analytics 4 measures how many people arrive, from where and what they do on your site. Google Tag Manager is a container for installing those tags without touching code each time.
- Map rankings (grid)
- Google Maps ranks businesses based on where the searcher is. That's why it's measured from several points across the city: each one shows where you rank. The average position counts «not listed» as 21, and the share of points in the top 3 (share of local voice) tells in what part of the city you show up in the map pack.

Tell me when it is ready
Free tool coming soon.
Leave your email and website: we will let you know when the tool is live. If you would rather have WITS review your site first, tell us what you want to achieve.
WhatsApp +52 56 6107 7931 · ventas@wits.mx