Glossary

AI & Customer Support Glossary

Clear definitions of the key terms in AI chatbots, customer support automation, and conversational AI.

Agentic AI

Agentic AI refers to artificial intelligence systems that can autonomously plan, reason, and execute multi-step actions to achieve a specified goal. Unlike traditional AI that responds to single prompts, agentic AI breaks down complex tasks, uses tools, makes decisions, and adapts its approach based on intermediate results.

AI Agent

An AI agent is a software system that perceives its environment through inputs (messages, data, events), reasons about goals and available actions, and acts autonomously to achieve those goals. AI agents combine language understanding with tool use, memory, and decision-making to operate with minimal human oversight.

AI Hallucination

AI hallucination is a phenomenon where a large language model generates text that is fluent, confident, and plausible-sounding but factually incorrect, fabricated, or unsupported by any source data. The model "hallucinates" information that does not exist in its training data or the provided context.

Average Handle Time (AHT)

Average Handle Time (AHT) is a customer support metric that measures the average total duration of a customer interaction from start to resolution. It includes active conversation time, hold or wait time, and any post-interaction wrap-up work (notes, ticket updates, follow-up actions).

Chatbot

A chatbot is a software application designed to simulate human conversation through text or voice. Chatbots range from simple rule-based programs that follow pre-defined scripts to sophisticated AI-powered systems that understand natural language and generate contextual responses.

Context Window

A context window is the maximum number of tokens that a large language model can process in a single request, encompassing all input (system prompt, conversation history, retrieved context) and output (the generated response). It represents the model's effective working memory.

Conversational AI

Conversational AI is a category of artificial intelligence that enables software to understand, process, and respond to human language in a natural, dialog-based format. It combines natural language processing (NLP), machine learning, and dialog management to power chatbots, virtual assistants, and voice interfaces.

CSAT (Customer Satisfaction Score)

CSAT (Customer Satisfaction Score) is a metric that measures how satisfied customers are with a specific interaction, product, or experience. It is typically measured by asking customers to rate their experience on a scale (1-5 or 1-10) immediately after an interaction.

Embedding

An embedding is a dense numerical vector (array of numbers) that represents the semantic meaning of a piece of text. Embedding models convert words, sentences, or documents into fixed-size vectors in high-dimensional space, where semantically similar texts are positioned close together.

Fine-Tuning

Fine-tuning is the process of taking a pre-trained large language model and further training it on a smaller, domain-specific dataset to specialize its behavior, knowledge, or output style. The model weights are updated to reflect the new training data, creating a customized version of the base model.

First Contact Resolution (FCR)

First Contact Resolution (FCR) is a customer support metric that measures the percentage of customer issues fully resolved during the first interaction — without requiring follow-up contacts, callbacks, escalations, or additional tickets. Also called first-call resolution or one-touch resolution.

First Response Time (FRT)

First Response Time (FRT) is a customer support metric that measures the elapsed time between when a customer submits a support request and when they receive the first meaningful response. It is one of the most important indicators of support responsiveness.

Human Handoff

Human handoff (also called human takeover, agent escalation, or live agent transfer) is the process of transferring an ongoing chatbot conversation to a human support agent. The agent receives the full conversation history so the customer does not need to repeat themselves.

Hybrid Search

Hybrid search is a retrieval method that combines semantic search (vector/embedding-based) with lexical search (keyword/BM25-based) to find relevant information. By merging both approaches, hybrid search achieves higher accuracy than either method alone.

Intent Classification

Intent classification is the process of analyzing a user message and categorizing it into a predefined intent — the action or information the user is seeking. For example, classifying "I want to cancel my subscription" as a "cancellation" intent and "How much is the Pro plan?" as a "pricing inquiry" intent.

Knowledge Base

A knowledge base is an organized collection of information — typically help articles, FAQs, guides, and documentation — that enables customers to find answers independently. In modern support, knowledge bases also serve as the training data for AI chatbots.

Large Language Model (LLM)

A Large Language Model (LLM) is a type of AI model trained on enormous amounts of text data to understand and generate human language. Examples include OpenAI GPT-5, Anthropic Claude 4.5, Google Gemini 3, and Meta Llama. LLMs power modern chatbots, writing assistants, and conversational AI.

Live Chat

Live chat is a communication channel that enables real-time text-based conversations between customers and human support agents through a website or app widget. Unlike email (asynchronous) or phone (voice), live chat provides immediate text-based support.

Natural Language Processing (NLP)

Natural Language Processing (NLP) is a branch of artificial intelligence that focuses on enabling computers to understand, interpret, and generate human language. It combines computational linguistics with machine learning to process text and speech.

NPS (Net Promoter Score)

Net Promoter Score (NPS) is a customer loyalty metric that measures how likely customers are to recommend your product or service to others. Customers rate their likelihood on a 0-10 scale and are classified as Promoters (9-10), Passives (7-8), or Detractors (0-6). NPS = % Promoters minus % Detractors.

Omnichannel Support

Omnichannel support is a customer service strategy that provides a seamless, unified experience across all communication channels — live chat, email, phone, social media, SMS, messaging apps, and self-service portals. Customer context and conversation history persist across channels, so customers never have to repeat themselves.

Prompt Engineering

Prompt engineering is the practice of designing, structuring, and refining the instructions (prompts) given to large language models to elicit accurate, relevant, and well-formatted responses. It encompasses system prompts, user prompt templates, few-shot examples, and output constraints.

Retrieval-Augmented Generation (RAG)

Retrieval-Augmented Generation (RAG) is an AI architecture that enhances large language model responses by first retrieving relevant information from a knowledge source, then using that information to generate accurate, grounded answers. Instead of relying solely on trained knowledge, RAG systems search your documentation in real time.

Semantic Search

Semantic search is a search technique that finds information based on the meaning and intent behind a query rather than matching exact keywords. It uses natural language understanding and vector embeddings to identify content that is conceptually relevant, even when the query and the content share no common words.

Sentiment Analysis

Sentiment analysis is an NLP technique that identifies the emotional tone (positive, negative, neutral, or specific emotions like frustration or urgency) in text. In customer support, it helps teams prioritize and respond appropriately to customer messages.

SLA (Service Level Agreement)

A Service Level Agreement (SLA) is a formal commitment between a service provider and a customer that defines measurable service standards — including response times, resolution times, uptime guarantees, and escalation procedures. In customer support, SLAs set expectations for how quickly and effectively issues will be handled.

Ticket Deflection

Ticket deflection occurs when a customer resolves their question or issue without creating a support ticket or contacting a human agent. This typically happens through self-service (knowledge base, FAQ) or AI chatbot resolution.

Token

A token is the fundamental unit of text that large language models process. Tokens are fragments of words, whole words, or punctuation marks that the model reads and generates. In English, one token is roughly 3/4 of a word — so 100 words is approximately 130-140 tokens.

Vector Search

Vector search is a method of finding information based on semantic meaning rather than exact keyword matches. It works by converting text into numerical representations (vectors/embeddings) and finding the most similar vectors in a database. Also called semantic search or similarity search.

Webhook

A webhook is an HTTP callback mechanism that automatically sends data from one system to another when a specific event occurs. Instead of continuously polling an API for updates, the source system pushes data to a configured URL in real time. Webhooks are the backbone of real-time integrations between SaaS applications.