The vocabulary, with the disagreements left in.
38 terms across search, AI discovery, attribution, revenue operations and web experience. Where the industry has no settled definition — and for 3 of these it does not — the entry says so instead of inventing a tidy boundary.
Reference
Search and discovery
15 termsBreadcrumbList schema
BreadcrumbList markup describes a page’s position in the site hierarchy as an ordered list. Google uses it to replace the raw URL in a result with a readable trail, which is one of the few rich results that still reliably renders.
Read the entryCanonical URL
A canonical URL is the address a site nominates as the authoritative version of a page when several URLs serve equivalent content. It is declared with a rel="canonical" link element, and it is a strong hint to search engines rather than a directive they are obliged to follow.
Read the entryCrawl budget
Crawl budget is the number of URLs a search engine will fetch from a site in a given period, set by how much load the server can absorb and how much demand the engine has for the site’s content. It becomes a practical constraint mainly on large sites.
Read the entryE-E-A-T
E-E-A-T stands for experience, expertise, authoritativeness and trustworthiness — the framework Google’s human quality raters use to assess page and site quality. Google states it is not itself a ranking factor; it describes what the systems are being tuned to approximate, and it matters most on topics that affect money or wellbeing.
Read the entryFAQ schema
FAQ schema (schema.org FAQPage) marks up question-and-answer content so machines can read the pairs. It is still a valid schema type, but Google discontinued the FAQ rich result for nearly all sites starting in May 2026, with phased removal from Search Console tooling through that August.
Read the entryFeatured snippet
A featured snippet is a short extract Google promotes above the standard results to answer a query directly, drawn from a page already ranking for it. Formats include a paragraph, an ordered or unordered list, and a table, and the format Google chooses follows the shape of the query.
Read the entryGoogle Business Profile (GBP)
Google Business Profile is the free listing that makes a business eligible for the map pack and the knowledge panel in local results. It requires verification, and it requires either a physical address customers can visit or a declared service area for businesses that travel to customers.
Read the entryhreflang
hreflang is an annotation that tells search engines which language and regional variant a page is intended for, so the right version is served to the right audience. Annotations must be reciprocal — every page in a set references every other, including itself.
Read the entryInternational SEO
International SEO is the work of making a site discoverable across multiple countries or languages: choosing a URL structure, signalling language and region, researching demand in each market’s own vocabulary, and adapting content to local buying behaviour and regulation.
Read the entryLocal search
Local search covers the results Google shows when a query has geographic intent — most visibly the map pack of three business listings above the organic results. Ranking there depends heavily on proximity, a verified business profile and consistent location data, none of which a remote business can supply.
Read the entryNAP consistency
NAP consistency means a business’s name, address and phone number appear identically everywhere they are published — the website, the Google Business Profile, directories and data aggregators. Inconsistent records make it harder for search engines to resolve them to one entity.
Read the entryPeople Also Ask (PAA)
People Also Ask is the expandable block of related questions Google shows within results, each opening to a short extracted answer and a source link. The questions are generated from real related queries, which makes the block a usable source of genuine subheadings.
Read the entrySearch engine optimization (SEO)
Search engine optimization is the practice of making a site easier for search engines to crawl, index, understand and rank, so that pages appear for queries the business can actually serve. It covers technical access, information architecture, content quality, internal linking, authority signals and the measurement that tells you which of those moved.
Read the entryStructured data
Structured data is machine-readable markup — usually JSON-LD using the schema.org vocabulary — that labels what a page’s content means: this is an article, this is its author, this is a service and its provider. Google confirms it is not a ranking factor; its documented job is enabling rich results and helping systems identify entities.
Read the entryYMYL
YMYL — "your money or your life" — is Google’s label for topics where inaccurate information could damage someone’s health, financial stability, safety or wellbeing. Pages on these topics are held to a higher evidence and trust standard in the quality rater guidelines.
Read the entry
AI and answer engines
8 termsAgentic workflow
An agentic workflow is an automation where a language model completes several connected steps toward a goal — researching, classifying, drafting, routing, reporting — deciding the order within defined rules, rather than executing a fixed script. Human approval points are part of the design, not an afterthought.
Read the entryAI Overviews
AI Overviews are Google-generated summaries shown above traditional results for some queries, assembled from pages in the ordinary search index and shown with links to the sources used. Eligibility requires that a page is already indexable and snippet-eligible in normal Search.
Read the entryAnswer engine optimization (AEO)
ContestedAnswer engine optimization is structuring content so a direct answer can be lifted cleanly into a featured snippet, a People Also Ask entry, a voice response or an AI summary. It predates the current AI wave — the original concern was voice assistants — and overlaps heavily with both SEO and GEO.
Read the entryGenerative engine optimization (GEO)
ContestedGenerative engine optimization is the practice of improving how a source is used by AI systems that synthesise answers rather than list links. The term comes from a 2023 academic paper (arXiv:2311.09735, later accepted at KDD 2024) that measured how often a document’s content was reflected in a generated answer.
Read the entryLLMO, AISO and GSO
ContestedLLMO, AISO and GSO are vendor terms for largely the same work as GEO and AEO: making content that AI systems can read, trust and reference accurately. No academic or platform-official definition separates them, and industry sources contradict each other on where the boundaries fall.
Read the entryllms.txt
llms.txt is a proposed convention: a Markdown file at a site’s root listing its important pages for AI systems. Adoption has grown quickly, but current measurement shows AI crawlers largely ignore it and fetch HTML directly instead.
Read the entryPrompt injection
Prompt injection is an attack where instructions hidden in content an AI system reads — a web page, an email, a document, a form submission — are treated by the model as commands rather than data. It is the main security concern for any agent that ingests untrusted input.
Read the entryRetrieval-augmented generation (RAG)
Retrieval-augmented generation is an architecture where a language model answers using documents fetched at query time rather than relying only on training data. The system splits sources into chunks, embeds and ranks them, places the best matches into the model’s context, and generates an answer from those.
Read the entry
Measurement and attribution
7 termsConsent mode
Consent mode is Google’s mechanism for adjusting tag behaviour according to a visitor’s consent choices, sending cookieless pings when consent is denied so that conversions can be modelled rather than simply lost. Its output is estimated, and reporting distinguishes observed from modelled data.
Read the entryGA4 key events
Key events are the GA4 events a property has marked as significant — the renamed successor to "conversions" in that interface. They drive reporting and can be imported into advertising platforms, and their counts may include modelled as well as observed data.
Read the entryIncrementality testing
Incrementality testing measures what a marketing activity actually caused, by comparing a group exposed to it against a comparable group that was not. Geographic holdouts, audience splits and time-based tests are the common designs, and the output is a causal estimate rather than an attributed share.
Read the entryMarketing mix modelling (MMM)
Marketing mix modelling uses aggregate historical data — spend, sales, seasonality, price, external factors — to estimate each channel’s contribution statistically. Because it works on aggregates rather than individual journeys, it is unaffected by cookie loss and consent restrictions.
Read the entryMulti-touch attribution (MTA)
Multi-touch attribution distributes credit for a conversion across the marketing touchpoints that preceded it, using a rule (first touch, last touch, linear, time decay) or a data-driven model. It describes observed sequence; it does not establish that any touchpoint caused the outcome.
Read the entryOffline conversion import
Offline conversion import sends outcomes that happen outside the website — a qualified lead, a booked job, closed revenue — back to the advertising platform, matched to the original click. It lets bidding optimise toward revenue rather than form fills.
Read the entryPerformance Max (PMax)
Performance Max is a Google Ads campaign type that serves across Search, Display, YouTube, Discover, Gmail and Maps from one budget, with placement, audience and creative decisions automated. It reports far less channel-level detail than the campaign types it replaced.
Read the entry
Revenue operations
5 termsLead routing
Lead routing is the logic that decides which person or team receives an inbound lead, based on territory, product, company size, source or availability. It runs automatically in most CRMs, and the rules tend to outlive the organisational structure that justified them.
Read the entryLead scoring
Lead scoring assigns a numeric value to a lead based on fit attributes (industry, size, role) and behaviour (pages viewed, emails opened, forms submitted), to prioritise follow-up. Its accuracy depends entirely on whether the model was built from outcomes rather than assumptions.
Read the entryPipeline velocity
Pipeline velocity estimates how quickly revenue moves through a sales process, combining the number of qualified opportunities, average deal value, win rate and average cycle length into a single rate-of-revenue figure.
Read the entryRevenue operations (RevOps)
Revenue operations is the function that aligns marketing, sales and customer success around one set of definitions, one data model and one measurement system. Its work is mostly at the seams — handoffs, routing rules, stage definitions, field hygiene and reporting that all three teams accept.
Read the entrySpeed to lead
Speed to lead is the elapsed time between an inbound enquiry arriving and a human making genuine contact. It is one of the few operational metrics where the direction of the effect is not seriously disputed — contact rates fall sharply as the interval grows.
Read the entry
Web experience
3 termsCore Web Vitals
Core Web Vitals are Google’s three user-experience metrics: Largest Contentful Paint (loading), Interaction to Next Paint (responsiveness) and Cumulative Layout Shift (visual stability). They are a confirmed ranking signal, and independent analysis indicates they act as a tie-breaker among comparable pages rather than a primary driver.
Read the entryLargest Contentful Paint (LCP)
Largest Contentful Paint measures when the largest visible element in the viewport finishes rendering — usually a hero image, a video poster or a large text block. Google’s threshold for "good" is 2.5 seconds at the 75th percentile of real visits.
Read the entrySearch experience optimization (SXO)
Search experience optimization joins ranking work to what happens after the click — page experience, clarity, task completion and conversion. It is the most stable term in the acronym cluster because it describes a genuine scope difference rather than a new surface, and it predates the generative-AI wave.
Read the entry
Corrections
Think a definition here is wrong?
Send the term, the part you disagree with, and the best source you have. Definitions in a moving field go stale, and a corrected entry is more useful than a confident one.
Continue with context
Choose the next useful path.
Go deeper into the evidence, compare the connected work, or bring us the constraint you are trying to resolve.
Questions
How engagements actually run
What starts an engagement, what we decline to guarantee, who owns the accounts, and where we work.
02Research
Go past the definition
Long-form guides on the same subjects, each with visible sources, dates and stated limitations.
03Delivery
See which of this we run
The services behind the vocabulary, with scope, prerequisites and failure modes named.
04Next step
Bring us a specific problem
Definitions are the easy part. The useful conversation starts with your constraint.