SEO & AI Glossary

Web pages that provide little unique value — through minimal text, duplicated information, or no original insight — and are susceptible to Google quality penalties.

Data collected by an external entity that has no direct relationship with the user — typically purchased from data brokers or ad platforms for audience targeting.

Content and positioning that establishes an individual or brand as a leading authority and innovative thinker in their industry.

The practice of optimising TikTok content to rank in TikTok's search results and in-app discovery feeds.

The time elapsed between a browser sending an HTTP request and receiving the first byte of a server response.

An attribution model that gives more credit to marketing touchpoints closer to the conversion, on the assumption that recency indicates higher influence on the purchase decision.

The process of breaking text into smaller units—tokens—that an AI model can process and understand.

The basic units of text that AI language models process — roughly corresponding to word fragments — used to measure input and output length and compute API costs.

A language model's ability to invoke external functions, APIs, or services during a conversation to take actions or fetch live data.

A content architecture model with one pillar page covering a broad topic and multiple cluster pages covering related subtopics, all interlinked.

The perceived depth and breadth of expertise a website demonstrates on a subject area, influencing how search engines rank its content.

A structured plan of all topics, subtopics, and content pieces needed to comprehensively cover a subject area and establish topical authority.

The total revenue opportunity available if a product achieved 100% market share of all potential buyers in its target category.

A low-quality or manipulative inbound link that can harm a website's search rankings or trigger a Google penalty.

The dataset used to teach a machine learning model, consisting of examples from which the model learns patterns and relationships.

A search intent type where the user is ready to take an action—typically to make a purchase or complete a conversion.

A technique where a model pre-trained on one task is adapted for a different but related task, saving time and data.

A deep learning architecture that processes sequences of data using self-attention, forming the basis of most modern LLMs.

Majestic's metric measuring the quality of links pointing to a URL, based on proximity to a curated set of trustworthy seed sites.

Design and content elements on a website that reduce visitor scepticism and increase confidence in a brand's credibility and reliability.