Customer support & AI glossary

An AI system that understands customer questions in natural language and resolves them autonomously, escalating to a human when needed.

A technique where an AI retrieves relevant passages from a trusted knowledge base and uses them to generate accurate, grounded answers.

Technology that lets software hold natural, back-and-forth conversations with people using language understanding and generation.

A chatbot follows pre-built rules and flows; an AI agent understands language and resolves open-ended questions autonomously.

Using AI and rules to handle repetitive support work — answering common questions, routing, tagging and escalating — with less manual effort.

AI that helps human agents in real time by drafting replies, surfacing knowledge and suggesting next steps inside the inbox.

Identifying what a customer is trying to accomplish from their message, so the system can route or answer appropriately.

Resolving customer questions through self-service or AI before they become a human-handled ticket.

When an AI confidently generates false or unsupported information as if it were fact.

An AI model trained on vast amounts of text that can understand and generate human-like language.

AI that creates new content — answers, summaries, replies — rather than only classifying or predicting.

The field of AI focused on enabling computers to understand and work with human language.

Automatically detecting the emotional tone — positive, negative or neutral — of a customer's message.

Search that understands the meaning behind a query, not just matching keywords.

Rules and safeguards that keep an AI's responses safe, accurate and on-policy.

Software that simulates conversation to answer questions or complete tasks — from simple rules to full AI.

Support that unifies every channel — chat, email, WhatsApp, social — into one connected experience with shared context.

Software that centralises customer requests into tickets or conversations so a team can track, assign and resolve them.

A structured library of help articles and answers that customers and AI can search to resolve questions.

Real-time messaging between customers and support, usually via a widget on a website or app.

Saved reply templates, often with dynamic variables, that agents insert to answer common questions quickly and consistently.

A commitment to respond to or resolve customer requests within a defined time, tracked and enforced by the support system.

Handing a conversation to a more senior agent, specialist team, or human — with full context — when it needs more than the current handler can provide.

A single inbox where a whole team can view, assign and reply to customer messages together.

Letting customers find answers themselves via help centers, FAQs and AI — without contacting an agent.

Reaching out to customers before they hit a problem — with guidance, updates or timely offers.

A tracked record of a customer request, from creation through to resolution.

Organising support into levels (tier 1, 2, 3) by complexity, escalating harder issues upward.

The overall impression a customer forms from every interaction with your business.

Guiding new customers to their first success with a product or service.

The official interface that lets businesses send and receive WhatsApp messages at scale through support software.

The embeddable chat box on a website or app that lets visitors start a conversation instantly.

Providing support on social channels like Instagram and Facebook Messenger where customers already are.

Businesses communicating with customers through messaging apps like WhatsApp and Messenger.

Support and messaging delivered inside a mobile or web app, in context.

A metric that measures how satisfied customers are with a support interaction, usually via a short post-conversation survey.

How long a customer waits for the first human or AI reply after reaching out.

The average time an agent spends handling a conversation from start to resolution.

A loyalty metric measuring how likely customers are to recommend you, on a 0–10 scale.

The share of issues resolved in a single interaction, without any follow-ups.

The average total support cost divided by the number of conversations handled.

A metric measuring how easy it was for a customer to get their issue resolved.

The percentage of customers who stop using your product or service in a given period.

The share of support conversations that are actually resolved, not just responded to.

Keeping existing customers over time — the flip side of churn.