WIT BOTS PLATFORM WIT Bots platformAI-powered Conversationalinterfaces

Why are bots important?

Reduce Operational Costs

Messaging bots can automatically answer between 20 and 30% of tier-1 customer questions. Along with a good end-user experience, bots can considerably reduce the OPEX costs of Contact Centers.

Stay Relevant

Every day, 4.1 billion people are using messaging Apps. With the latest advances in artificial intelligence, humans are starting to interact with machines and software systems. This is the best way to stay relevant in the customer’s eyes.

Create New Revenue Streams

A carrier-grade Bots platform can provide value-added conversational services to consumer brands, productivity bots to the enterprise market, and realize the new paradigm of conversational commerce for the retail industry. This can be a new source of monetization for Mobile Operators.

More Convenient to use

People can use messaging Bots to interact 24/7 with companies, providing a more focused and tailored experience than websites and custom mobile Apps.

Messaging as a Platform

The new Messaging as a Platform paradigm is obtained by the use of three main software components:

Messenger as a Platform Diagram
WIT A2P Gateway

Enables a simple and secure network API for accessing the mobile operator network

WIT Bots Platform

Allows management of 3rd party bots and creating bots for customer service, enterprise productivity and conversational commerce.

Key benefits
  • Reduce cost from operator customer service
  • Sets the Mobile Operator right in the middle of the revenue stream in new services provided by messaging bots
  • Allows a controlled exposure of community to global service providers

Deploying Scenario

The WIT Bot Platform can be deployed in a hosted or cloud environment.

WIT Bots Deployment Scenario

Key features

WIT Bot Gateway is the module responsible for enabling bots into messaging channels, as well as controlling bots using rules to avoid frustration and spamming.


Presents each Bot as a contact user and maps the MSISDN or SIP URI to the Bot name.


Allows the user to find Bots by searching in the messaging App.

Content Adaptation

Adapts messaging format based on the channel: FB Messenger, Skype, RCS, SMS, etc.

3rd Party Bot API

Similar to the Facebook Messenger API, it allows a frictionless API for connecting third-party Bot providers.

Sentiment Analysis

Detects sentiment of frustration, diverting more complex conversations to human assistants.

Bot Firewall

Avoids spamming actions by defining rules for push messages.


Provides key performance indicators for analysis, such as: engaged users, lost users, engaged sessions and conversations per session.

Messaging Channels

The available messaging channels are RCS, Skype, Facebook Messenger, Telegram, Slack, Amazon Alexa and SMS.


Key features

The WIT Bot Engine is a Visual bot development and runtime environment that allows a comprehensive development and management for your bot.

Bot Templates

Pre-built templates for customer service, enterprise productivity and retail Bots.

Workflow Designer

Conversation flows based on BPMN (Business Process Management Notation).

Conversation Designer

Conversation-authoring, context and multi-lingual support.


Unleash the power of Natural Language Processing and Machine Learning from industry-leading providers.

Workflow Engine Layout

Digital Brain

The Bot Digital Brain is responsible for understand the user’s query and returning an answer according to its pre-loaded knowledge base. It is composed by two sub-systems:

Workflow Engine Layout

Enables the comprehension of Natural Language, using integrated technology from industry leading partners

Holds the trained sets of intents and entities, and the relation to workflow events they will fire once activated

Manages and maintains the set of trained intents and entities

Knowledge base

Uses Machine Learning, based on the most reliable and performing Artificial Intelligence technology.

Computes and stores a knowledge base, using a data classifier, that learn from text files, using semantic data extraction technology.

Allows knowledge recalibration when a human performs a correction into its neural outcome.

Conversational tools in WIT Bot Engine

In order to create good bots, great tools are required. WIT Bot engine offers tools to develop bots with high retention and engaging rates.

Conversation Engagement

Pre-built 10k sentences to allow chitchat engagement conversation, with the bot’s most frequent asked questions (Bot profile, Emotions, Greetings, Jokes, Tastes, Smiles reaction)

Multi Conversation Sets

Configure and deploy multi bot personalities, in several languages.

Pre-built Telco domain specific dictionary

Allows advanced natural language processing thanks to a set of synonyms to telco dictionary

Emotional Response

Sentiment analysis calculated in each interaction, to enrich conversation and emotional response

Steer back

Actionable behavior that brings back the user when conversation starts to go out-of-flow.

wit-bots-transactional-conversations copy
Transactional Conversations

Make you conversational flow transactional, and force a commit or rollback before user changes subject.

Fall back Strategy

Allows cascade search for a good answer when trigger activation fails

Customer Support Bots

Automatic Answer of Tier-1 Questions

Using a knowledge base powered by Machine Learning, the software Bot can easily learn from available Q&A files.

Answers are contextualized with user’s product subscription.

When the Bot is not able to answer, or is frustrating the end-user, the conversation is automatically diverted to a human assistant.

Offers Troubleshooting Guidance

Conversation is done step-by-step to help users diagnose the problem.

Built-in troubleshooting decision tree to guide the user.

Providing Account Information

Data is obtained from integrated data sources.

Configurable triggers to enable up-selling.

Customer Support Bots

Human assistance

A Web Interface for human agents to achieve the best management of Bot conversations. It avoids frustration of the end-user and enables the Bot to evolve and learn from these situations

Human assistance
Conversation Correction

The human assistant can then review the conversation and engage the correct flow for the answer.

Offers Troubleshooting Guidance

Conversation flow is analyzed in real-time through an AI engine of sentiment analysis.

If there is any frustration in the conversation, it is diverted to a human assistant.


When human correction occurs, the bot will learn from that, increasing its knowledge for similar situations in the future.

Retail bots

Conversational Commerce

Enables conversational commerce using connectors to online product catalogs.

Retail bots

Enterprise bots

Bots add value by increasing the productivity of workers with easy access to enterprise information and services. Enterprise bots can be deployed in several use-cases:

IT Services
Document Management
Meeting Assistance
Travel Assistance
Conference Calls
Expense Reporting
HR Info

Want to know more about our bots? Check out a detailed PDF or get in touch with us.

Check the PDF Contact Us