Reference

AI, SaaS, and engine optimization glossary

A working reference of 177 terms across artificial intelligence, machine learning, SaaS architecture, custom software, web development, search and engine optimization, and SDVOSB federal contracting. Each term below has a stable URL you can link to or cite directly.

Jump to a letter

Browse the glossary

8 A B C D E F G H I J K L M N O P Q R S T U V W Y

8

8(a) Business Development Program

SBA program for socially and economically disadvantaged small businesses. 9-year program providing federal contracting access, mentorship, and training. Stricter eligibility than other set-asides.

A

ADA Title III

Americans with Disabilities Act, Title III applied to commercial websites. Lacks specific technical standards but DOJ accepts WCAG AA as a benchmark. ADA lawsuits common.

Agentic AI

AI systems that operate semi-autonomously through tool use, planning, and reasoning. Enables multi-step workflows like 'research this topic and write a report' or 'fix this bug across the codebase'.

Ahrefs

A commercial SEO platform competitor to Semrush. Excellent backlink index. Keyword research and content gap analysis, priced from about 99 to 1000 dollars a month.

AI Agent

An LLM-powered system that can take actions through tools (web search, file systems, APIs, code execution) to accomplish multi-step tasks autonomously. Includes Claude Code, ChatGPT operator-style systems, and open frameworks like LangGraph.

AI Citation

Being cited as a source by an AI engine (ChatGPT, Claude, Perplexity, Gemini) when answering user questions. The new center of gravity in search visibility.

AI Optimization (AIO)

Structuring a site to be cited by ChatGPT, Claude, Perplexity, Gemini, and other LLM-based search engines. Includes llms.txt, factual density, citation worthiness, and embedding distinctiveness.

Anchor Text

The visible clickable text of a link. Descriptive anchor text helps users and search engines understand the destination. Avoid 'click here' and 'read more.'

Answer Engine Optimization (AEO)

Optimizing for direct answer engines (featured snippets, People Also Ask, knowledge panels) and voice assistants (Alexa, Siri, Google Assistant). Often emphasizes Q&A structure, schema, and entity authority.

Anthropic

AI safety lab founded in 2021. Creator of the Claude family (Claude 3.5 Sonnet, Claude 3.7 Sonnet, Claude 4.x Opus/Sonnet/Haiku) and Constitutional AI methodology. API at anthropic.com.

Apple Silicon

Apple's ARM-based processors (M1, M2, M3, M4). High performance per watt. Increasingly used for local AI inference because of unified memory architecture. Powers MEGAMIND.

Application Programming Interface (API)

A defined way for software to talk to other software. Typically HTTP/REST or GraphQL. APIs let you integrate services like Stripe billing, Twilio SMS, OpenAI, or your own internal tools.

Artificial Intelligence (AI)

Computer systems that perform tasks typically requiring human intelligence, including language understanding, decision making, perception, and reasoning. In small business contexts, AI usually refers to the use of large language models like GPT-4 or Claude to automate or enhance specific workflows.

Astro

A multi-framework static site generator. Renders content as plain HTML by default, hydrates only what needs interactivity. Excellent for content-heavy sites and blogs.

Attention Mechanism

A method in neural networks for weighting how strongly each part of an input influences each part of an output. Allows transformers to focus on relevant context when generating each token.

Automation Script

Code that performs a repetitive task without human intervention, such as file processing, data sync, or scheduled report generation. Often written in Python, Node.js, or Bash.

B

Backlink

A link from another website pointing to yours. Quality (authority, relevance) matters more than quantity. Toxic backlinks (spam, paid) can harm rankings.

Bing Webmaster Tools

Microsoft's webmaster tool for Bing search. Provides similar data to GSC. Supports IndexNow for immediate URL submission. Required since Microsoft Copilot is Bing-integrated.

BM25

Best Matching 25, a classic keyword based ranking function used in search engines. Powers Elasticsearch and OpenSearch. Often combined with semantic search in hybrid retrieval.

Schema markup for site navigation showing the path from homepage to current page. Renders as a hierarchy in Google search results.

C

Canonical URL

The official, preferred URL for a piece of content. Set with <link rel='canonical'>. Critical for preventing duplicate content issues. Self-referencing canonicals are required on canonical pages.

Capability Statement

A one to two page document summarizing a federal contractor's core competencies, past performance, certifications, NAICS codes, and contact info. Required for any federal pursuit.

CCPA / CPRA

California Consumer Privacy Act / California Privacy Rights Act. Gives California residents rights over their personal data similar to GDPR. CPRA expanded CCPA in 2023.

Chain of Thought (CoT)

A prompting technique where the LLM is asked to reason step-by-step before producing an answer. Improves accuracy on math, logic, and complex reasoning. Many modern models do CoT automatically when needed.

Chunking

Splitting documents into smaller pieces before embedding. Strategies: fixed-size, sentence-based, paragraph-based, semantic, recursive. Chunk size affects retrieval quality dramatically.

CI/CD

Continuous Integration / Continuous Deployment. Automated testing and deployment pipelines triggered by code changes. Tools: GitHub Actions, GitLab CI, CircleCI, Vercel.

Citation (Local)

Any mention of a business's NAP across the web. Structured citations: directory listings (Yelp, BBB, Yellowpages). Unstructured: news articles, blog mentions. Both improve local authority.

ClaudeBot / anthropic-ai

Anthropic's web crawlers. ClaudeBot is the search/citation crawler; anthropic-ai is the broader user-agent. Allow both for Claude citations.

Cloudflare Workers

Serverless JavaScript/TypeScript functions that run on Cloudflare's edge network. Fast cold start, global deployment. Powers many production AI integrations.

CMMC

Cybersecurity Maturity Model Certification, a DoD requirement for contractors handling controlled unclassified information (CUI). Levels 1 through 5, with Level 2 most common for SDVOSB AI and SaaS work.

Content Audit

A systematic review of all content on a site for quality, performance, and relevance. Identifies pages to keep, improve, consolidate, or remove.

Content Delivery Network (CDN)

A global network of servers caching static assets close to users. Reduces latency, increases availability. Cloudflare, Fastly, AWS CloudFront, Bunny CDN, Akamai are major providers.

Context Window

The maximum number of tokens an LLM can process in one prompt. GPT-4-Turbo has 128K tokens; Claude 3.5 Sonnet has 200K; Claude with extended thinking can reach 1M. Larger windows allow more documents in RAG.

Core Update

A broad change to Google's ranking algorithm released several times per year. Recent: March 2024 (largest ever), August 2024, November 2024, December 2024, March 2025. Impacts can take weeks to settle.

Core Web Vitals

Three Google performance metrics: LCP (Largest Contentful Paint, target < 2.5s), INP (Interaction to Next Paint, target < 200ms), CLS (Cumulative Layout Shift, target < 0.1). Required for good rankings.

Cron / Scheduled Tasks

A time-based job scheduler in Unix-like systems. Runs scripts at specified intervals (every hour, daily at 3am, etc.). Cloud equivalents: AWS EventBridge, GitHub Actions schedule, Vercel Cron.

Cumulative Layout Shift (CLS)

The amount of unexpected layout shift while a page loads. Target under 0.1. Improved with explicit image dimensions, reserved space for ads, and avoiding late-loading content.

Custom Software

Software built specifically for one organization's needs rather than purchased off the shelf. Examples: internal tools, custom dashboards, automation scripts, data pipelines, custom APIs.

D

Data Pipeline

A series of steps that move and transform data from source to destination. ETL (Extract-Transform-Load) or ELT (Extract-Load-Transform). Tools: Airflow, Prefect, Dagster, dbt.

Docker

Containerization platform. Packages applications with their dependencies into portable containers. Industry standard for deployment.

E

E-E-A-T

Experience, Expertise, Authoritativeness, and Trustworthiness. Google's framework for evaluating content sources. Critical for YMYL queries. Implemented through author bylines, credentials, sameAs, citations, and editorial signals.

Embedding

A numerical vector representation of text where similar meanings produce similar vectors. Used in semantic search to find relevant documents by meaning rather than keyword match. Common dimensions are 384, 768, 1024, 1536.

Entity Salience

How important an entity is within a piece of content based on NLP analysis. Per-page entity engineering helps Google understand what each page is primarily about.

ETL (Extract Transform Load)

The process of pulling data from source systems, transforming it (cleaning, joining, enriching), and loading it into a destination database or warehouse.

F

FAQPage Schema

Schema for marking up question-and-answer content. Can produce rich results in search. Use only when the page genuinely answers questions.

Featured Snippet

A direct answer extracted from a webpage and shown at the top of Google results. Position zero. Triggered by question queries, achieved through clear, structured Q&A content.

FedRAMP

Federal Risk and Authorization Management Program, a security standard for cloud services used by federal agencies. Low, Moderate, and High impact levels. A long, expensive certification.

Few-Shot Prompting

Providing the LLM with several worked examples in the prompt before asking it to handle a new case. Helps the model match a desired output format, style, or reasoning pattern.

Fine Tuning

Training an existing LLM on additional data to specialize its behavior or knowledge. More expensive than RAG and harder to update. Often misused, since for most use cases RAG is faster, cheaper, and easier to maintain.

Freemium

A pricing model with a free tier and paid upgrades. Common in SaaS to drive viral adoption. Risk: users stay on free tier; signal quality drops.

Function Calling / Tool Use

A capability where an LLM outputs structured JSON specifying which tool/function to invoke with which arguments. Enables agents to interact with external systems. OpenAI calls it function calling, Anthropic calls it tool use.

G

General Data Protection Regulation, an EU law on personal data. Requires lawful basis for processing, data subject rights such as access and deletion, and breach notification within 72 hours.

Generative Engine Optimization (GEO)

Optimizing for generative AI search results, specifically Google AI Overviews and similar synthesized answer surfaces. A subset of AIO focused on Google's SGE derived experiences.

Git

Distributed version control system. Standard for source code management. GitHub, GitLab, Bitbucket are hosted providers.

GitHub

The largest hosted Git provider. Owned by Microsoft. Provides repos, pull requests, issues, Actions (CI/CD), Pages (static hosting), and Copilot.

Google AI Overview

Google's generative answer feature appearing above organic results. Synthesizes information from multiple sources with citations. A major shift in search visibility, and appearing here is increasingly critical.

Google Business Profile (GBP)

Formerly Google My Business. The free listing system for local businesses on Google Search and Maps. Critical for local SEO. Verified ownership unlocks insights and review responses.

Google Gemini

Google's family of multimodal LLMs. Powers Google AI Overviews, Bard (now Gemini), and Vertex AI. Strong at long context (1M+ tokens) and code.

Google Search Console (GSC)

Google's free webmaster tool. Provides indexing status, crawl errors, query data, click-through rate, and Core Web Vitals. Required for any serious SEO effort.

Google-Extended

A separate user-agent Google uses for Bard/Gemini training. Distinct from Googlebot (which is for Search). Allow Google-Extended to be considered for Gemini training.

GPTBot

OpenAI's web crawler used to gather data for ChatGPT. Allow it in robots.txt to be considered for ChatGPT citations. Blocks reduce LLM training inclusion.

GraphQL

A query language for APIs developed by Meta. Clients request exactly the data they need. More flexible than REST for complex frontends but more complex to operate.

GSA / GSA Schedule

General Services Administration, which runs federal wide acquisition programs. The GSA Schedule (formerly Schedule 70 for IT) is a multiple award contract pre vetting vendors for federal agencies.

H

Hallucination

When an LLM generates plausible-sounding but factually incorrect output. Occurs because LLMs predict statistically likely tokens, not verify facts. Mitigated by RAG, citation requirements, and fact-checking layers.

Hand Coded Website

A website built directly in HTML, CSS, and JavaScript without a CMS or builder. Faster, lighter, and more controllable than WordPress or Squarespace. Tradeoff: requires a developer for content updates.

Headless CMS

A content management system that exposes content via API but doesn't render it. Frontend is decoupled, can be built with any framework. Examples: Sanity, Contentful, Strapi, Payload.

Helpful Content System (HCS)

Google's framework for evaluating whether content was created primarily for users (helpful) or for search engines (low-quality). Asks Who/How/Why questions.

HIPAA

Health Insurance Portability and Accountability Act, a US law protecting health information. Requires Business Associate Agreements (BAA), encryption, audit logs, and breach notification.

Horizontal SaaS

SaaS used across many industries (CRM, project management, file storage). Larger market but harder to differentiate. Examples: Salesforce, Notion, Slack.

HowTo Schema

Schema for marking up step-by-step instructional content. Each step gets a HowToStep type with name, text, and optional image. Often used with totalTime and estimatedCost.

hreflang

A meta tag specifying the language and region of a page. Enables Google to serve the right version to the right audience. Common values: en-US, es-US, en-GB, x-default.

HTTP/2

The 2015 update to HTTP. Multiplexed streams, header compression, server push. Faster than HTTP/1.1 for sites with many resources. Required for modern SEO.

HTTP/3 / QUIC

The 2022 update to HTTP. Built on QUIC (UDP-based) instead of TCP. Faster connection setup, better mobile performance. Adoption growing across modern hosting.

HUBZone

Historically Underutilized Business Zone, an SBA program for businesses in distressed areas. A set aside category for federal contracts.

Hugging Face

Open AI hub hosting hundreds of thousands of models, datasets, and demos. Provides Transformers library, Inference Endpoints, and Spaces. Standard repository for open source ML.

Hybrid Search

Combining keyword search (BM25) with semantic search (embeddings) to get the best of both. Often paired with a reranker for final ranking. Standard for production RAG.

I

Incremental Static Regeneration (ISR)

A Next.js feature that combines static generation with on-demand revalidation. Pages are static but can be regenerated when content changes. Best of static + dynamic.

IndexNow

An open protocol for instant URL submission to search engines. Bing, Yandex, and Naver participate. Submit URL on publish/update; engines respond within minutes.

Information Gain

Google Patent US 11,995,114 B2. A measure of how much new information a piece of content provides relative to existing search results. Original research, contrarian analysis, first-hand experience score high.

Interaction to Next Paint (INP)

The time from a user interaction (click, tap, key press) to the next paint update. Target under 200ms. Improved with smaller JavaScript bundles, breaking up long tasks, and using Web Workers.

Internal Link

A link from one page on a site to another. Distributes PageRank, provides crawl paths, and helps thematic clustering. Hub-and-spoke architecture is the dominant pattern.

Internal Tool

Software used inside an organization (not by customers). Common types: admin dashboards, data entry forms, reporting tools, automation panels. Often built fast with React + a backend or with Retool.

J

Jamstack

JavaScript + APIs + Markup. An architecture pattern where static HTML is served from CDN, dynamic features come from JavaScript and external APIs. Faster, more secure, and easier to scale than traditional server-rendered apps.

JSON Web Token (JWT)

A compact, URL-safe token format used for authentication. Contains claims signed by a secret. Common in API authentication and stateless session management.

JSON-LD

JavaScript Object Notation for Linked Data, the format Google recommends for schema.org markup. Embedded in script tags of type application/ld+json.

K

Keyword Cannibalization

When multiple pages on the same site target the same keyword, splitting authority and confusing ranking signals. Resolved by consolidating, redirecting, or differentiating.

Knowledge Graph

Google's database of entities (people, places, things, concepts) and their relationships. Powers knowledge panels, sitelinks, and entity-based search. Wikidata is the open equivalent.

Kubernetes (K8s)

Container orchestration platform. Manages containers across clusters of machines. Required for large-scale SaaS but overkill for small projects.

L

Large Language Model (LLM)

Neural networks trained on massive text corpora that generate natural language. GPT-4, Claude, Gemini, and Llama are examples. They can answer questions, write code, summarize, translate, and reason about text.

Largest Contentful Paint (LCP)

The time at which the largest content element above the fold finishes loading. Target under 2.5 seconds. Improved with image optimization, preloading, and reducing render-blocking resources.

Link Audit

Analysis of inbound and outbound links for authority, toxicity, and pattern detection. Identifies link spam, paid link patterns, and disavow candidates.

Linux

Open source Unix-like operating system. Powers most servers and cloud infrastructure. Common distros: Debian, Ubuntu, RHEL, Alpine.

LLM Optimization (LLMO)

A near-synonym for AIO. Sometimes used to specifically describe optimizing for inclusion in LLM training data versus retrieval citations.

llms.txt

An emerging standard file at the root of a domain providing structured site context for AI engines. Contains site purpose, key URLs, and citation guidance. Companion file llms-full.txt provides full markdown context.

Local Pack / Map Pack

The 3-result map result Google shows for local queries. Highly visible. Driven by GBP, proximity, relevance, and prominence.

Local SEO

Optimization for queries with local intent (near me, in [city]). Includes Google Business Profile, NAP consistency, local citations, reviews, and local landing pages.

LoRA / QLoRA

Low Rank Adaptation, a lightweight fine tuning technique that trains small adapter weights rather than the whole model. QLoRA quantizes the base model first. Enables fine tuning on consumer GPUs.

M

Machine Learning (ML)

A subset of AI where systems learn patterns from data rather than being explicitly programmed. Includes supervised, unsupervised, and reinforcement learning. Modern LLMs are an extension of ML using deep neural networks.

Meta Llama

Meta's open weights LLM family (Llama 3, 3.1, 3.2, 3.3). Released under permissive license for commercial use. Often run self-hosted via Ollama, vLLM, or HuggingFace TGI.

Meta Robots / X-Robots-Tag

Tags controlling crawler behavior per page. index/noindex, follow/nofollow, max-snippet, max-image-preview. Set in <meta name='robots'> or HTTP X-Robots-Tag header.

Minimum Viable Product (MVP)

The smallest version of a product that delivers value and validates core hypotheses. Built fast to get user feedback. Not low quality, but low scope with focused execution.

Mistral AI

French AI lab building open and proprietary LLMs. Mistral 7B, Mixtral 8x7B, Mistral Large, Codestral. Strong at code and multilingual tasks.

Model Context Protocol (MCP)

An open standard from Anthropic for connecting LLMs to external tools, data sources, and services. Lets models call standardized servers for filesystem access, database queries, web search, etc. Enables portable agent ecosystems.

Multi-Tenant Architecture

A SaaS pattern where one application serves many customers (tenants) with logical isolation. Cheaper to operate than dedicated instances. Most modern SaaS is multi-tenant.

N

NAICS Code

North American Industry Classification System, six digit codes identifying business activities. Federal contracts list eligible NAICS codes. Common AI codes include 541511, 541512, and 541519.

NAP (Name, Address, Phone)

Core business identity for local SEO. Must be consistent across the website, GBP, citations (Yelp, Bing Places, etc.), and structured data. Inconsistencies hurt local pack rankings.

Netlify

A deployment platform for Jamstack sites. Provides build pipelines, edge functions, forms, identity. Direct competitor to Vercel.

Next.js

A React framework for production-grade web apps. Provides server-side rendering (SSR), static generation (SSG), incremental static regeneration (ISR), API routes, and server components. Industry standard for React SaaS in 2026.

Nginx

A high-performance web server and reverse proxy. Handles HTTPS termination, load balancing, caching, gzip/brotli compression. Standard for production self-hosted sites.

NIST 800-53

National Institute of Standards and Technology Special Publication 800 53. Security and privacy controls for federal systems. Foundation for FedRAMP, FISMA, and many other frameworks.

O

OAuth 2.0

An authorization standard. Lets users grant a SaaS app access to their account at another service without sharing their password. Used by Google Sign In, GitHub Login, etc.

Ollama

A local LLM runner for macOS, Linux, and Windows. Easy local inference with Llama, Mistral, Phi, and other open models. Built on llama.cpp.

OpenAI

AI lab founded in 2015. Creator of GPT series (GPT-3.5, GPT-4, GPT-4-Turbo, GPT-4o), DALL-E, Whisper, and o1/o3 reasoning models. Provides API access at openai.com.

Orphan Page

A page with no internal links pointing to it. Hard for Google to discover and rank. Found and fixed during regular link audits.

OWASP LLM Top 10

OWASP's list of the top 10 vulnerabilities in LLM applications: prompt injection, insecure output handling, training data poisoning, etc. Reference for AI security audits.

P

PCI DSS

Payment Card Industry Data Security Standard. Requirements for handling credit card data. Self-assessment for small merchants; full audit for large ones. Stripe and Paddle handle most PCI burden for SaaS.

PerplexityBot

Perplexity's web crawler. Required for Perplexity to cite the site. Blocking it eliminates the site from Perplexity's index.

pgvector

PostgreSQL extension adding vector search. Lets Postgres serve as a vector database. Excellent for small/medium RAG applications. Free, mature, simpler than dedicated vector databases.

PostgreSQL (Postgres)

Open source relational database. Industry standard for SaaS. Supports JSON, full text search, and pgvector for embeddings, making it a viable single database for AI applications.

Product Service Code (PSC)

Four-character codes used by federal agencies to classify what they're buying. Complement NAICS. Common AI/software PSCs: D307 (data processing), D316 (telecom), D317 (web design), R408 (program evaluation).

Product-Led Growth (PLG)

A go-to-market strategy where the product itself drives acquisition, expansion, and retention. Examples: Slack, Notion, Figma. Requires excellent self-service onboarding.

Prompt Engineering

The practice of writing inputs that elicit reliable, accurate outputs from LLMs. Includes system prompts, few-shot examples, role definitions, and structured output specifications.

Q

Quantization

Reducing the precision of model weights from 16-bit floating point to 8-bit, 4-bit, or even lower. Cuts memory and inference cost dramatically with modest accuracy loss. Common formats: GGUF, GPTQ, AWQ.

Queue / Background Job

A list of work items processed asynchronously. Decouples slow tasks (sending email, generating reports) from the user request. Common queues: Redis-based (Bull, Sidekiq), AWS SQS, Cloudflare Queues.

R

React

A JavaScript library for building user interfaces, created by Meta. Component-based, declarative. The most popular web UI library. Foundation for Next.js, Remix, and many others.

ReAct Pattern

Reasoning plus Acting. An agent loop where the LLM alternates between thinking about the problem and taking actions through tools. A foundational pattern for AI agents.

Redis / Valkey

In-memory data store. Used for caching, queues, pub/sub, rate limiting. Valkey is the BSD-licensed fork after Redis license change.

Request for Information (RFI)

A federal pre solicitation document gathering market intelligence. Not a request for proposals, informational only. Vendors respond with their capabilities and recommendations.

Request for Proposal (RFP)

A formal federal solicitation requesting detailed proposals from qualified vendors. Includes statement of work, evaluation criteria, and submission requirements.

Reranker / Cross Encoder

A second stage model that takes top K candidate documents from initial retrieval and reorders them for relevance. Cohere Rerank and BGE Reranker are common. Adds roughly 100 to 300 milliseconds but significantly improves quality.

REST API

Representational State Transfer, an API style using HTTP verbs (GET, POST, PUT, DELETE) and JSON. The most common API style today. Standard for SaaS integrations.

Retrieval Augmented Generation (RAG)

A technique where an LLM retrieves relevant documents from a vector database, then generates an answer grounded in those documents. RAG reduces hallucinations and lets LLMs answer with current, proprietary, or domain-specific information.

robots.txt

A file at the root of a domain telling search engines which paths to crawl and which to skip. Also lists sitemap URLs. AI bots like GPTBot, ClaudeBot, and PerplexityBot have specific rules.

S

SaaS Billing

Subscription-based payment processing. Stripe, Paddle, Lemon Squeezy, Chargebee handle recurring charges, prorations, plan changes, dunning, and revenue recognition.

SAM.gov

System for Award Management, the official US government registration site for federal contractors. Free registration. Required for any federal opportunity.

sameAs Property

A schema.org property linking an entity to its profiles on other authoritative sites (Wikidata, LinkedIn, GitHub, social profiles). One of the strongest entity authority signals.

SBIR / STTR

Small Business Innovation Research and Small Business Technology Transfer, federal programs funding small business R&D. Phase I, II, and III with increasing funding levels.

Schema.org / Structured Data

A vocabulary of types and properties for marking up entities on web pages. Implemented as JSON-LD in <script> tags. Enables rich results in Google Search and citations in AI engines.

SDVOSB

Service Disabled Veteran Owned Small Business, a federal certification (verified through CVE/VetCert) reserving certain federal contracts, called set asides, for service disabled veteran owned firms. A strong differentiator in federal opportunities.

Search Engine Optimization (SEO)

The practice of structuring a website so it ranks well in search engines like Google and Bing. Covers technical (crawling, indexing, schema), content (relevance, quality, depth), and authority (links, mentions, brand).

Search Generative Experience (SGE)

Google's earlier name for AI Overviews. Discontinued as a public name, now called AI Overviews. Same underlying technology.

Search Quality Rater Guidelines (SQRG)

The 175-page document Google's human raters use to evaluate sites. Public document outlining E-E-A-T, page quality, and needs-met assessment.

Section 508

US federal accessibility law requiring electronic information from federal agencies be accessible. References WCAG 2.0 AA (updated to 2.1 in 2018).

Semantic Search

Search that finds documents by meaning rather than keyword match. Powered by embeddings and vector databases. Returns 'similar in concept' rather than 'contains the same word'.

Semrush

A commercial SEO and competitive analysis platform. Keyword research, backlink data, competitor tracking. Industry standard but expensive, roughly 120 to 450 dollars a month.

Server-Side Rendering (SSR)

Rendering pages on the server when requested, then sending HTML to the browser. Best for personalized content and dynamic data. Tradeoff: higher server cost, slower TTFB than static.

Set-Aside Contract

A federal contract reserved for a specific small business category (8(a), SDVOSB, WOSB, HUBZone). Increases competitiveness for qualifying businesses.

Single-Tenant Architecture

Each customer gets their own dedicated application instance. More expensive but offers stronger isolation. Common in regulated industries (HIPAA, PCI) and government contracts.

sitemap.xml

An XML file listing all canonical URLs on a site, with optional lastmod, changefreq, and priority fields. Submitted to Google Search Console. Image and video sitemaps are separate variants.

SOC 2

AICPA security framework for SaaS vendors. Type I is a point in time review; Type II is a six to twelve month continuous audit. Increasingly required by enterprise B2B customers.

Software as a Service (SaaS)

Software delivered over the internet as a subscription rather than a one-time purchase. Hosted by the vendor, accessed via browser. Examples: Salesforce, HubSpot, Notion.

Sources Sought Notice

A federal pre-solicitation notice asking interested vendors to identify themselves. Helps the agency understand the supplier landscape and decide whether to use a set-aside.

SpeakableSpecification

Schema marking which parts of a page are suitable for voice/audio reading. Voice assistants and Google Assistant use this to extract spoken responses.

Static HTML

Plain HTML files served from disk or CDN. The original web architecture. Simple, fast, secure. Modern static site generators produce this output.

Static Site Generation (SSG)

Pre-rendering pages at build time. Fastest option since HTML is served directly from CDN. Limitation: can't show user-specific content without client-side fetch or revalidation.

Stripe

Payment processor and SaaS billing platform. Handles credit cards, ACH, recurring subscriptions, plan changes, invoicing, and tax collection. Industry standard for SaaS billing.

Svelte / SvelteKit

A JavaScript framework that compiles components at build time. Smaller bundle, faster runtime than React. SvelteKit is the full-stack framework. Growing alternative to Next.js.

System Prompt

The high-level instruction given to an LLM that defines its role, tone, constraints, and behavior. Sets the context for the entire conversation. Critical for production AI deployments.

T

Tailwind CSS

A utility-first CSS framework. Style components directly with classes (px-4, text-blue-500). Fast development, consistent design tokens. Industry standard since 2022.

Technical Audit

A review of crawling, indexing, performance, structured data, and on-page technical signals. Tools: Screaming Frog, Sitebulb, GSC, Lighthouse.

Temperature

A parameter controlling randomness in LLM outputs. Temperature 0 is deterministic (best for code and structured output). Temperature 0.7 to 1.0 is creative (best for brainstorming and varied phrasing).

Token

The unit LLMs process. A token is roughly three to four characters of English. The phrase 'small business AI' is about 4 tokens. LLM pricing and context limits are measured in tokens.

Top-P (Nucleus Sampling)

A sampling method where the LLM picks from the smallest set of tokens whose cumulative probability exceeds P. Top-P 0.9 is common and eliminates very low probability tokens while keeping diversity.

Transformer

The neural network architecture introduced in 2017 that powers modern LLMs. Uses attention mechanisms to weigh relationships between tokens. Most modern AI systems including GPT, Claude, and Llama are transformer-based.

TypeScript

A typed superset of JavaScript developed by Microsoft. Adds static type checking. Standard for production frontend and Node.js code in 2026.

U

Unique Entity ID (UEI)

A 12-character alphanumeric ID assigned by SAM.gov to identify federal contractors. Replaced the old DUNS number in 2022.

V

Vector Database

A database optimized for storing and querying high-dimensional vectors (embeddings). Examples: Pinecone, Qdrant, Weaviate, Chroma, Milvus, LanceDB, and pgvector for Postgres. Powers semantic search and RAG.

Vercel

A deployment platform for Next.js and other frontend frameworks. Provides edge deployment, serverless functions, preview deploys per branch. Created by the Next.js team.

Vertical SaaS

SaaS purpose-built for one industry (legal, healthcare, real estate, fitness). Deeper feature fit than horizontal SaaS but smaller TAM. Examples: ClubReady, ToadsAndDragons, Fitbox.

vLLM

High-throughput LLM inference engine. PagedAttention algorithm dramatically improves GPU utilization. Standard for production LLM serving.

VOSB

Veteran Owned Small Business, a broader category than SDVOSB that covers veteran owned businesses without the service disabled requirement. Used by VA contracts.

W

WCAG 2.2

Web Content Accessibility Guidelines version 2.2. International standard for web accessibility. AA conformance is the standard target. Required for Section 508 compliance and ADA defense.

Webhook

An HTTP callback. When event X happens in service A, A sends a POST to a URL you specify. Used for real time integration, for example Stripe sending an event when a customer pays.

Wikidata

The free, structured knowledge base used by Wikipedia and many AI engines. Each entity has a Q-ID (e.g., Q139901957 for Joseph Anady). Critical for entity authority and AI citation signals.

WOSB / EDWOSB

Women Owned Small Business / Economically Disadvantaged Women Owned Small Business. Set-aside category for women-owned businesses with specific revenue and ownership tests.

Y

YMYL (Your Money or Your Life)

Google's classification for queries affecting health, finances, safety, civic participation, or major life events. Held to higher quality standards. Requires stronger E-E-A-T signals.

Begin

See where a federation fits your work

If a term here describes a gap in how your organization runs AI today, the readiness assessment turns it into a real plan.

Start with an assessment Or start a conversation