Glossary of Terms
Chatbot
A computer program that simulates human conversation through voice or text commands. Primarily used for customer service, information gathering, and sales.
AI (Artificial Intelligence)
A field in computer science that deals with creating intelligent machines capable of performing tasks that typically require human intelligence.
Lead Generation
The process of creating consumer interest in a business's products or services with the goal of converting that interest into a sale.
Intent
The goal or objective of the user when contacting a chatbot. For example, "check order status" or "schedule a meeting".
Entity
A specific piece of information that the bot identifies and extracts from the user's request to fulfill the intent. For example, "order number", "date", or "product name".
Utterance
Everything the user writes or says to the chatbot. For example, "Hey, where's my package?" is an utterance whose intent is "check order status".
NLP (Natural Language Processing)
A branch of artificial intelligence that enables computers to understand, interpret, and generate human language (text and speech).
NLU (Natural Language Understanding)
A subfield of NLP that focuses on understanding the meaning and intent behind text or speech, including entity recognition and sentiment.
NLG (Natural Language Generation)
A subfield of NLP that enables a computer to generate readable and coherent human language text, i.e., to formulate the bot's responses.
Flow / Script
The predetermined sequence of steps, questions, and answers that the bot follows to guide the user to complete a task.
Fallback
The response the bot provides when it fails to understand the user's request or does not have an appropriate answer.
Live Chat Handoff
The process by which the chatbot smoothly transfers the conversation to a human service representative, usually when the problem is too complex for automated handling.
Quick Replies
Buttons with suggested answers that the user can click instead of typing, to speed up and simplify the conversation.
Carousel
A visual component that allows the bot to display multiple items (such as products or articles) in horizontally scrollable cards, each with an image, title, and buttons.
Machine Learning
A field in AI that enables systems to learn and improve from data and experience, without being explicitly programmed for every situation.
Training Data
A large set of examples (for instance, user utterances and their corresponding intents) used to "teach" the bot's AI model to recognize patterns.
LLM (Large Language Model)
An advanced AI model trained on massive amounts of text, capable of understanding and generating human language at a very high level, such as GPT models.
Containment Rate
A KPI metric representing the percentage of conversations that the chatbot successfully resolved in full, without the need to transfer to a human representative.
Goal Completion Rate
A KPI metric that measures the percentage of users who successfully completed a defined goal through the bot (for example, registering for a webinar).
Sentiment Analysis
An automated process for identifying and classifying the emotion expressed in text - positive, negative, or neutral. Helps understand user satisfaction.
Omnichannel
A strategy that provides a unified and continuous customer experience across all channels (website, WhatsApp, Messenger, physical store, etc.), with context preserved when moving between channels.
Bot Persona
The character and personality of the chatbot, including its speaking style (Tone of Voice), vocabulary, and behavior. Essential for creating a consistent experience identified with the brand.
API (Application Programming Interface)
An interface that allows different software systems to "talk" to each other. Chatbots use APIs to connect to CRM systems, inventory, and more.
Webhook
An automated mechanism that allows one system to send information in real-time to another system the moment a certain event occurs. For example, sending a lead from the bot to CRM immediately upon receipt.
A/B Testing
A method for comparing two versions of a component (for example, message wording) to determine which one yields better performance.
Proactive Bot
A chatbot that initiates a conversation with the user based on their behavior (for example, after 30 seconds on the pricing page), rather than waiting for the user to contact it.
Transcript
A complete record of the entire conversation that took place between the user and the chatbot, including timestamps.
Context
The information the bot "remembers" from previous parts of the conversation, enabling it to conduct a continuous and logical conversation.
Disambiguation
When the user's request can be interpreted as multiple intents, the bot will request clarification. For example: "Did you mean to cancel an order or cancel a meeting?".
Configuration
The process of setting up and customizing the chatbot, including script setup, integrations, and design.
CUI (Conversational User Interface)
An interface based on conversation (text or voice) instead of graphical components like buttons and menus (GUI).
Gateway
A software component that bridges between the chatbot platform and a specific communication channel, such as WhatsApp or Facebook Messenger.
Variables
Temporary "containers" for storing information during a conversation, such as the user's name or the last product selected. The information is deleted at the end of the conversation.
User Attributes
Permanent information details stored on the user's profile between conversations, such as purchase history, preferred language, or subscription status.
Segmentation
Dividing the user audience into groups based on shared attributes or behaviors, to enable sending personalized messages.
Broadcasting
Sending an initiated message to a large number of users at once, usually for marketing purposes or important updates.
Message Template
A message pre-approved by a channel like WhatsApp, allowing businesses to initiate a conversation with customers. Usually required for broadcast messages.
Integration
Connecting the chatbot to other software systems (such as CRM, ERP) to enable synchronization and real-time data retrieval.
Latency
The time it takes for the bot to respond to the user's request. Low latency is essential for a good user experience.
Scalability
The ability of the chatbot platform to handle an increasing number of users and simultaneous conversations without affecting performance.
Authentication
The process of verifying the user's identity, for example by sending a one-time code, to enable the bot to perform sensitive actions on their account.
SDK (Software Development Kit)
A collection of tools and code libraries that enables developers to embed and integrate chatbot capabilities into mobile applications or complex websites.
CSAT (Customer Satisfaction Score)
A common metric that measures customer satisfaction with a specific interaction, usually through a question like "How satisfied were you with the service?" on a scale of 1-5.
NPS (Net Promoter Score)
A metric that measures customer loyalty to a brand by asking "How likely are you to recommend us to a friend?" on a scale of 0-10.
Neural Network
A computational model inspired by the human brain, which forms the basis for deep learning technologies and enables bots to understand complex language.
Hybrid Bot
A chatbot that combines rule-based and scripted capabilities with AI-based capabilities for understanding free-form language.
Visual Builder
A tool for building conversation scripts using drag-and-drop of graphical components, without the need to write code.
Channel
The platform where interaction with the bot takes place, such as a website, WhatsApp, Facebook Messenger, Telegram, or Slack.
Rich Media
Bot responses that are not text-only, but also include images, videos, GIF files, audio files, or documents.
Keyword
A specific word or phrase that the bot is programmed to recognize and trigger an automatic response when it encounters it.
JSON (JavaScript Object Notation)
A lightweight text format for transferring data between systems, very common in chatbot API communication.
Automation
The execution of processes and tasks automatically without human intervention, such as sending order confirmation or answering common questions.
Ticketing System
A system (like Zendesk or Jira) for managing service requests. A bot can create a new ticket in such a system when it fails to resolve a problem.
UX (User Experience)
The overall experience of a person using a product, and in this case, the feeling and efficiency of the conversation with the chatbot.
Onboarding
The initial process in which the bot introduces itself to the user, explains what it can do, and guides them on how to use it.
Voice Chat
Interaction with a bot through speech, rather than typing. Common in telephony systems (IVR) and smart voice assistants.
Conversational Form
A way to collect information from the user through a series of questions in a conversation, instead of presenting them with a standard form with fields to fill out.
Session Duration
A KPI metric that measures the average duration of a conversation between a user and a bot.
GDPR (General Data Protection Regulation)
A strict European regulation that governs the collection and processing of personal information of European Union citizens. Relevant to any bot operating in Europe.
PII (Personally Identifiable Information)
Any information that can be used to identify a specific person, such as name, phone number, email address, or ID number.
Anonymization
The process of removing identifying information (PII) from data, to maintain user privacy and enable safe data analysis.
Dialog Engine
The core system of the chatbot, responsible for managing conversation flow, maintaining context, and deciding on the next response.
Confidence Score
A numerical score (usually between 0 and 1) representing the AI engine's level of certainty that it correctly identified the user's intent.
Multilingual
The ability of a chatbot to communicate with users in multiple languages, and often to automatically detect the user's language.
Speech Recognition
The technology that converts spoken language (voice) into written text, enabling the bot to understand voice commands.
Speech Synthesis
The technology that converts written text into artificial human voice, enabling the bot to "speak" its responses.
Roadmap
A strategic plan describing the future development of the chatbot, including planned features, improvements, and integrations.
CTA (Call to Action)
A clear instruction to the user to perform a desired action, such as a "Buy Now" button or "Sign Up Here".
Conversion Funnel
The path the user takes from the initial interaction with the bot to completing a desired goal (such as a purchase or registration).
Retraining
The periodic process of updating the bot's AI model with new data (real conversations) to improve its accuracy and performance.
Abandonment Rate
The percentage of users who leave the conversation with the bot before completing the task or goal.
Stop Word
A common word (like "the", "of", "I") that NLP engines often filter to focus on the more important keywords in a sentence.
Rule-Based Bot
A simple type of chatbot that follows a fixed script and responds only to specific words or commands defined in advance. Does not use AI for understanding free-form language.
Entity Validation
An automatic check that the information provided by the user matches the expected format. For example, verifying that an email address contains "@" or that a phone number contains only digits.
CRM (Customer Relationship Management)
A software system for managing all interactions and information related to customers. Bots connect to CRM to retrieve customer information and save new leads.
Conversational Sitemap
A visual diagram describing all intents, scripts, and relationships between them in a chatbot, similar to a website sitemap.
Deployment
The process of uploading the latest version of the chatbot to the production environment, making it available to end users.
Staging Environment
A copy of the production environment used for testing new changes and features before they go live, to prevent issues.
Widget
A small software component that can be embedded in a website. The chat widget is usually the small bubble in the corner of the screen that opens the conversation window.
Error Handling
How the bot responds to unexpected failures, such as API communication failure, to provide the user with as smooth an experience as possible.
Guided Conversation
A conversation style where the bot leads the user through quick questions and answers, rather than allowing free input, to ensure the conversation stays on the desired track.
Open-Ended Conversation
A conversation style that allows the user to type freely, relying on advanced NLU capabilities to understand their intent.
Encryption
Converting information into code to prevent unauthorized access. End-to-end encryption ensures that only the sender and recipient can read the messages.
Model
In AI, this is the file or computational structure created during the machine learning process, which actually performs tasks like intent recognition or sentiment analysis.
RPA (Robotic Process Automation)
Technology that uses "software robots" to perform repetitive tasks in computer systems. Can be integrated with chatbots to perform complex operations in the background.
Asynchronous Messaging
A type of communication (like WhatsApp or email) where participants do not have to be available at the same time, and the conversation can continue over hours or days.
IVR (Interactive Voice Response)
An automated telephony system that allows callers to navigate menus using keypad input or voice commands. Modern IVR is essentially a voice chatbot.
Heatmap
A visual tool in analytics that shows which intents or scripts are most popular in use with the bot.
Customer Journey
The entire set of interactions and experiences of a customer with a company over time. A chatbot can serve as an important touchpoint in this journey.
Community Building
Using a chatbot in channels like Discord or Telegram to manage communities, answer member questions, and share updates.
Acceptance Testing
A final testing stage where it is verified that the chatbot meets all the business and functional requirements defined in advance.
Brand Voice
The personality and values of the brand as expressed in written communication. The bot's persona should match the brand voice.
Domain
The specific knowledge area that the chatbot specializes in, for example, "banking", "fashion", or "technical software support".
Reserved Word
A word like "help" or "stop" that usually triggers a global function in the bot, regardless of the current stage in the conversation.
Intent Prediction
The ability of an AI model to predict the user's intent even before they finish typing, based on the beginning of the sentence and conversation context.
Node
In a visual builder interface, each box or component in the conversation flow diagram (such as sending a message, asking a question) is called a node.
Edge
In a visual builder interface, the line connecting two nodes and defining the transition between them in the conversation script.
Voice Cloning
AI technology that enables creating a synthetic voice very similar to a real person's voice, based on short voice samples.
Speaker Recognition
The ability of a voice system to identify who is speaking based on their unique "voiceprint", for security and personalization purposes.
FAQ (Frequently Asked Questions)
One of the most common uses for a chatbot is providing automatic answers to the organization's list of frequently asked questions.
Semantic Search
A type of search that understands the meaning and context of search words, not just exact word matching. Enables bots to find relevant answers in knowledge bases.
Dependency
A situation where one action in the bot (for example, checking inventory) depends on the success of another action (for example, identifying the requested product).
Knowledge Base
A structured information repository (like a collection of support articles) that the chatbot can search to find answers to user questions.
Data Bias
A situation where the bot's training data does not represent reality or includes prejudices, which may cause the bot to provide biased or unfair answers.
AI Governance
The set of rules, policies, and processes that an organization defines to ensure responsible, ethical, and safe use of artificial intelligence technologies like chatbots.
Infinite Loop
A bug in script design where the bot repeats the same question or action over and over without the ability to exit the state.
State
The current stage or location of the conversation within a defined script. State management is critical for complex conversations.
Multi-turn Conversation
A conversation that includes multiple exchanges (requests and responses) between the user and bot, as opposed to a single question-answer conversation.
Conversational Commerce
The use of conversational platforms (like chatbots) to enable customers to discover, choose, and purchase products directly within the conversation.
Confirmation
A stage in the conversation where the bot repeats the information it understood from the user and requests their confirmation before performing an action. For example: "I understood you want to order a pizza with olives. Should I confirm?".
CLV (Customer Lifetime Value)
An estimate of the total revenue a business can expect to earn from a customer over the entire period of their relationship. Bots can increase CLV by improving service and loyalty.
Agent
Another term for a chatbot or a human service representative handling conversations.
Unified Inbox
An interface that centralizes all conversations from all channels (WhatsApp, website, Messenger) in one place, enabling representatives to manage them efficiently.
Unsupervised Learning
A type of machine learning where the model learns to identify patterns in data without labels or explicit guidance. Useful for identifying new topics in conversations.
Supervised Learning
The most common type of bot training, where the model learns from labeled data (for example, examples of utterances and their correct intent).
CES (Customer Effort Score)
A metric that checks how much effort is required from the customer to resolve their problem. Effective bots aim to reduce customer effort.
Operator
The person or team responsible for monitoring, managing, and improving the chatbot.
Tagging
Adding labels to conversations or users to classify them. For example, tagging a conversation as "complaint" or a user as "VIP customer".
Typing Indicator
The animation of three dots (...) that appears when the bot is "preparing" its response, signaling to the user that the system is working.
Welcome Message
The first message the user receives when they start a conversation with the chatbot.
End of Conversation
A point in the script where the bot recognizes that the task is completed, usually accompanied by a farewell message and a request for feedback.
Product Roadmap
A document describing the vision, direction, and planned progress of a software product (like a chatbot platform) over time.
Regression Testing
A testing process designed to ensure that new changes in the bot's code have not broken existing functionality that worked correctly in the past.