Business messaging glossary WhatsApp, customer service and conversational AI terms
Short, vendor-neutral definitions of the terms you meet when you set up WhatsApp, a shared inbox or an AI assistant, each with a link to the place where MonoChat puts it to work.
The vocabulary of the WhatsApp Business Platform and other customer messaging channels.
Broadcast (bulk messaging)
Sending the same message, usually personalised with variables, to many contacts at once. On WhatsApp, broadcasts through the Business Platform use approved templates and should go only to opted-in contacts. Replies arrive as individual conversations.
A company approved by Meta to give businesses access to the WhatsApp Business Platform. A BSP typically handles onboarding, number registration and billing, and often adds its own inbox or automation tools on top.
A message made of several horizontally scrollable cards, each with its own image or video, text and buttons. It is used to show a set of products, offers or options in a single message.
A link, button, QR code or ad that opens a messaging conversation with a business when tapped. Click-to-chat ads on social platforms send people straight into WhatsApp, Instagram or Messenger instead of a landing page.
The limited period after a customer’s most recent message during which a business may reply with free-form messages on WhatsApp. Each new customer message restarts the window. Once it closes, the business needs an approved template message to write again.
A single workspace where messages from several channels, such as WhatsApp, Instagram, web chat and SMS, appear together. Agents answer from one screen, and the history of a customer follows them across channels.
A customer’s explicit permission to receive messages from a business on a given channel. Messaging platforms and privacy laws expect opt-in before promotional messages are sent. A clear way to opt out belongs with it.
A button under a message that lets the customer answer with one tap instead of typing. Quick replies keep answers structured, which makes them easy to use in automated flows. The same term is also used for saved responses that agents insert into a chat.
A pre-written WhatsApp message format that Meta reviews and approves before it can be sent. Templates are required when a business starts a conversation or writes outside the customer service window. They can contain variables, media and buttons, and fall into categories such as marketing, utility and authentication.
A chat window embedded in a website or app, usually opened from a floating button. Visitors can ask questions without leaving the page, and the conversation reaches the same team or assistant as other channels.
The programmatic interface that lets a business send and receive WhatsApp messages from its own software instead of the phone app. It supports multiple agents, automation and integrations on a single number. It has no user interface of its own, so it is used through a platform or custom code.
Meta’s current name for its WhatsApp offering for medium and large businesses. The Cloud API is the version hosted by Meta, so a business does not have to run its own servers to use it. In everyday use, “WhatsApp Business API” and “WhatsApp Business Platform” refer to the same thing.
A setup in which the same phone number works in the WhatsApp Business app and on the WhatsApp Business Platform at the same time. A business can keep chatting from the phone while a platform adds a shared inbox, automation and integrations.
A WhatsApp feature for building structured, form-like screens that open inside a chat. Customers can pick dates, fill in fields or choose options without leaving WhatsApp. Typical uses are bookings, lead forms and surveys.
Terms teams use to organise, measure and improve customer conversations.
Conversational commerce
Selling through messaging conversations: customers discover products, ask questions, order and pay inside a chat. It combines product catalogues, payment links and order updates with human or AI assistance.
Software that stores customer records, deals and interaction history in one place. Connecting a CRM to messaging lets agents see who they are talking to and keeps conversations attached to the right contact.
A measure of how satisfied customers are with a specific interaction, usually collected with a short rating question right after it ends. It is reported as the share of positive answers or as an average score.
The time between a customer’s first message and the first reply from the business. It is one of the most common customer service metrics, because long waits at the start of a conversation strongly affect satisfaction.
The moment a conversation passes from a bot or AI assistant to a human agent, or from one agent to another. A good handover carries the conversation history and collected details so the customer does not have to repeat themselves.
A loyalty metric based on one question: how likely a customer is to recommend the company, on a scale from 0 to 10. The score is the percentage of promoters (9–10) minus the percentage of detractors (0–6).
The rules that decide which agent or team receives an incoming conversation. Routing can be based on channel, language, topic, customer value, working hours or agent workload.
A commitment to a measurable level of service, for example a maximum first response or resolution time. Support teams track SLAs per conversation and escalate those at risk of breaching them.
Rules and visual flows that carry out repeatable steps without manual work, such as sending a reminder, tagging a contact or assigning a conversation. A workflow starts from a trigger, checks conditions and runs actions.
How chatbots, AI agents and the models behind them work.
AI agent
An AI system that does more than answer: it plans steps, uses tools and completes tasks toward a goal, such as booking an appointment or checking an order. In customer service, AI agents work within the permissions and instructions a business gives them.
Software that answers people in a chat automatically. Rule-based chatbots follow predefined menus and decision trees, while AI chatbots use language models to understand free text and write their own replies.
The ability of a language model to request a defined function, such as “check order status”, with structured parameters instead of only writing text. The application runs the function and returns the result, which lets an assistant act on real systems.
An answer from a language model that sounds confident but is factually wrong or unsupported by its sources. Grounding answers in a knowledge base, limiting scope and handing uncertain cases to a person reduce the risk.
The organised collection of documents, FAQs, policies and product information that a support team or AI assistant answers from. Its accuracy and freshness largely determine the quality of automated answers.
A neural network trained on very large amounts of text to understand and generate language. LLMs power modern chatbots and AI agents. They are offered by several providers and differ in quality, speed, cost and supported languages.
An open protocol that standardises how AI applications connect to external tools and data sources. A system exposes its capabilities through an MCP server once, and any compatible AI client can discover and use them.
A technique in which the system first retrieves relevant passages from a company’s own content and then gives them to a language model to write the answer. It grounds replies in current, specific information and reduces invented answers.
The standing instructions given to a language model before a conversation starts. It defines the assistant’s role, tone, boundaries and what to do in specific situations, and applies to every message in the conversation.
A database that stores text as numerical vectors (embeddings) and finds entries by similarity of meaning rather than exact keywords. It is the usual storage layer behind RAG and semantic search.
The technical building blocks that connect messaging to the rest of your systems.
API (Application Programming Interface)
A defined way for one piece of software to request data or actions from another. Messaging platforms use APIs to connect channels, CRMs, payment providers and internal systems.
A phone system that greets callers with recorded or synthesised prompts and routes them by keypad input or speech. Modern IVRs can hand a caller to an agent, a voice bot or a messaging channel.
A short code that is valid for a single login or transaction and expires quickly. Businesses deliver OTPs by SMS, WhatsApp or voice to verify that a person controls a phone number.
The signalling protocol used to set up, manage and end voice and video calls over the internet. A SIP trunk connects a business phone system or contact centre software to the public telephone network.
An HTTP request that one system sends to another automatically when an event happens, such as a new message or a completed payment. Webhooks let systems react in real time without repeatedly asking for updates.