Request for Information Template

Conversational AI
RFI & RFP Template

Based on 100+ Conversational AI RFIs, we've compiled the latest questions and requirements, including Large Language Models, RAG and more.



RFI / RFP Template for Conversational AI in 2024

Get a head start with our Conversational AI RFI template. You will find 60+ answers for our low-code AI platform for contact center automation.

Take a Peak at the Latest RFI Questions for AI

Product Overview

Question Answer Link
What's your Product's name? Cognigy.AI - AI Agents for your Business

What is the main focus of your product?

Cognigy.AI focuses on providing AI Agents powered by Conversational and Generative AI, empowering enterprises to deliver exceptional customer service that is instant, personalized, in any language, and on any channel. These AI Agents are pre-trained with industry-specific skills, can speak over 100 languages, and integrate seamlessly with enterprise systems to understand your service knowledge, services, and processes, delivering on-demand contextual answers. They also assist human agents in real time, providing proactive help, knowledge lookups, next best actions, sentiment analysis, call wrap-up, and more.

What are the key features of your product that differentiate it from competitors?

1. Seamless integration of Generative AI to harness the power of LLMs and build better, faster bots, deliver more human service and more.
2. Easy to use graphical conversation flow editor allows conversations to be built without code, no technical skills required
3. Built-in NLP with industry leading accuracy for intent recognition
4. Open source extensions repository with drag and drop installation (pre-built connectors for SalesForce, ServiceNow, Microsoft Office365 and many more)
5. SaaS, Dedicated hosting or On-Premises deployment options
6. Seamless live agent handover for contact center integration
7. Rapid project implementation time frames
8. Machine-learning agnostic (built-in connectors for Watson, Dialogflow, LUIS and open NLP pipelines)

What are the target industry sectors that your product specializes in?

Cognigy.AI solutions are technologically industry agnostic.  However, Cognigy provides pre-trained AI Agents specialized in a variety of specific industries such as: Airlines & Travel, Automotive, Financial Services, Healthcare, Insurance, Retail/eCommerce, Telecommunications and Utilities. 

Cognigy.AI's intent collections and pre-built solutions for various industries enable a quick start into a successful conversational AI project. 

What are the primary chatbot building features offered by your product?

Cognigy offers two options: pre-trained and ready to go AI Agents, built by Cognigy that are ready to launch as well as a platform option to build and customize them yourself. 

For builders, Cognigy.AI is a highly-flexible conversational automation platform to build advanced, integrated enterpriseAI Agents for customer and employee service on chat and voice channels. 

Cognigy.AI enables business users to create human-like automated conversations with the graphical conversation editor that harnesses the power of the Cognigy NLU in an easy to use format without the need for writing code.

What skills does a bot developer require to use your product?

Cognigy.AI is built for non-technical users and offers a comprehensive selection of conversational building, NLU training and third party integration tools that require minimal programming knowledge to use. This even goes for back-end integrations, as many have been pre-built and therefore do not require custom development. Cognigy.AI also offers a fully featured developer CLI for technical users to build Cognigy.AI projects in code from their preferred IDE.

Generative AI / LLMs

Question Answer Link
Does your solution incorporate and/or integrate with LLMs (Generative AI) such as ChatGPT, OpenAI and others? Yes. Firstly, Cognigy.AI offers out-of-the-box LLM integrations for a variety of vendors including OpenAI (including on Azure), Anthropic, Google, Aleph Alpha as well as an open API to easily add any others.  LLM features are integrated throughout the entire product for:

-LLM enhanced live interactions 
-Improved NLU understanding and entity extraction
-Transcription, sentiment analysis and analytics
-Enhanced language output
-Auto-generate intents
-Auto-generate lexicons
-Create flows and suggest next steps (nodes)
-Improve Agent Assistance via AI Copilot
-Knowledge understanding and generation
-and more


Does your solution offer integrations with external LLM vendors? If so, please list the vendors. Yes. Microsoft Azure OpenAI, OpenAI, Anthropic, Google, Aleph Alpha


Does your solution offer integrations for multiple LLM models from the same vendor?

Yes, this is called Multi-Model LLM Orchestration. Tis allows the configuration of multiple large language models in a virtual agent. Each model can be used for the purpose it’s best suited for, depending on speed, cost, and quality of responses for a given use case.

Do your LLM integrations include pre-engineered prompts to ensure consistent performance?


Is it possible to use an in-house or custom built/tuned LLM? 



Conversation Design

Question Answer Link

Does Cognigy offer prebuilt/pretrained bots or virtual assistants?

Yes, Cognigy offers prebuilt/pretrained bots, known as AI Agents, which come with industry-specific skills.  They are designed to quickly deployed and integrated seamlessly with enterprise systems across various channels. Cognigy's professional services team is responsible for building and ongoing optimization.

How fast can an AI Agent be developed and deployed?

Depending on the complexity and use case, it could take anywhere from an hour to several weeks. Cognigy can build and deploy for you, typically within several weeks. Alternatively, you also have the option to build your own.

How does the product interpret user messages and manage output dialogs?

The product interprets user messages through its integrated NLU engine which uses pre-trained language models trained on curated data in over 100 languages. The NLU also supports out-of-the-box prebuilt entity recognition for 28 languages. Output dialogs are managed by the chat- and voicebots built in the product, which support a wide variety of different channels including digital chat-channels, voice-channels, and smart assistants such as Amazon Alexa. Customers can also connect their custom messengers and platforms via generic REST, Webhook, and endpoints.

Do you provide any pre-built agents for immediate deployment?

Download the full list of responses with over 30 additional answers.

Natural Language Understanding

Question Answer Link

Does your product include an NLU component and how is it applied?

Yes, Cognigy includes its own Natural Language Understanding (NLU) component. This NLU component is a crucial part of the conversational AI capabilities of the Cognigy.AI platform. It enables virtual agents to understand and interpret natural language input from users, which is essential for processing and responding to user queries effectively. The NLU component in Cognigy.AI is designed to handle various aspects of language understanding, including intent recognition and entity extraction, which are vital for creating responsive and interactive conversational experiences. Additionally, Cognigy.AI supports integration with third-party NLU engines through NLU Connectors, allowing users to connect with other NLU providers like Google Dialogflow and IBM Watson Assistant if needed.

Is it possible to use an external NLU service with your product?

"Yes, Cognigy.AI supports the use of external Natural Language Understanding (NLU) services. It includes out-of-the-box connectors for several leading NLU providers, such as Amazon Lex, Amazon Alexa Skills Kit, Microsoft LUIS, IBM Watson, and Google DialogFlow. When these connectors are used, Cognigy's own NLU engine is overridden, and the results from the connected NLU service are used instead, allowing for seamless integration and utilization of external NLU capabilities within Cognigy.AI flows.


How effective is the NLU for recognizing user intents?

The Cognigy NLU has proven to outperform key industry competitors for user intent recognition. Cognigy.AI allows non-technical users to apply advanced machine learning algorithms to trained user example sentences via a simple and easy to understand user interface that enables user intent recognition with market leading accuracy. More information regarding benchmark testing can be found on our website.

Is the NLU able to build on its understanding and learn from customer interactions?

Cognigy.AI Agents constantly improve their understanding by adding misunderstood phrases to their existing machine learning intent records when users confirm this is appropriate. The Cognigy.AI Intent Trainer also allows platform users to review misunderstood inputs to improve the accuracy of the Machine Learing Intents manually. It is easy to add an input text from a user as an example sentence to a selected ML Intent, or add it to the Reject Intent of a Flow.

Can external NLU services be used with your product?

Cognigy.AI offers full support and customization options for integrating an external NLU service such as Watson, MS LUIS and Dialog flow with Cognigy.AI virtual agents. The NLU connectors feature provides the integration channel for connecting to an NLU provider. Once set up, the external NLU is made interchangable throughout the platform alongside the built-in Cognigy NLU.

Which languages does the product support?

Cognigy.AI supports a wide range of languages for Natural Language Understanding (NLU). It is pre-trained with data from over 100 languages to support intent recognition and key phrase detection. For 28 of the most common languages, Cognigy.AI provides prebuilt entities that allow automatic processing of inputs like dates, currencies, and others specific to a language that is defined in the flow.

Do you provide tools to assess the quality of training models?

Download the full list of responses with over 30 additional answers.

Contact Center Features

Question Answer Link

Is it possible to handover conversations to human contact center agents?

Cognigy.AI provides a built-in agent handover node which can be used if a request from a user is an edge case that is just too specific to be recognized by the virtual agent. The bot can offer the user to be forwarded to a support agent, a real human, that can intercept the conversation and help the customer manually without the necessity of changing channels.

Can the product support both chat and voice conversations?

Cognigy.AI provides a voice channel connector called the the Cognigy Voice Gateway. This feature allows any virtual agent flows to be deployed to voice channels by incorprating text to speech and speech to text services inline with message input and output. The same conversation flow can be deployed to any number of output channels enabling simple conversation management in chat or vocie from a single flow.

Can the conversation history and data context of a conversation be sent to human agents?

Cognigy.AI virtual agent conversations can be customized to transfer a summary of the conversation transcript, conversation data and user profile information during agent handover process. This is typically handled by accessing and summarizing the Odata analytics records into a formatted transcript.

What contact center solutions are supported by the product?

Download the full list of responses with over 30 additional answers.

Integration & Customization

Question Answer Link

How easy is it to integrate your product with external systems?

Cognigy.AI is an all-inclusive platform package delivering a highly flexible and customizable enterprise ready solution. Built-in nodes allow external integrations to be created without code, with the option of adding custom features to the user interface. In addition, all Cognigy.AI functions are accessible externally via API to be activated or created by external systems.

What types of integrations are possible for retrieving and storing information and executing processes?

Cognigy.AI fully supports back-end integrations with built-in https nodes allowing API calls to send or retrieve information within conversation flows. Similar native interactions are possible with SQL and mongo db with extended interactions possible by installing any of the open source extensions available on the Cognigy github repository.

Is it possible to customize the user interface by adding customer specific tooling?

The Cognigy.AI platform is extremely flexible due to an open-source extension framework. This is a key differentiator and allows for rapid development of fully-integrated bots with pre-packaged custom flow nodes that can be installed to each agent. Extensions customize flow nodes that are written in typescript and allow the appearance (image, colour and title) of the node to be customized in addition to the settings available in the editor menu and of the actual function that the node performs. An open-source library of existing extensions is available on the Cognigy github page.

Do you provide any plug-in features to connect directly with existing back-end services?

Download the full list of responses with over 30 additional answers.

Analytics & Logging

Question Answer Link

Does the product offer reporting features for monitoring performance and usage?

Cognigy.AI contains a built-in project analytics dashboard which displays key chatbot metrics to users within the platform. Conversation data is also available via the Cognigy Odata feed which can be used to import live conversation data to external visualisation tools such as Tableau and Power BI.

How does the product allow users to investigate issues and are there support options provided?

Cognigy.AI provides full developer logging that monitors all comunications to or from the platform. Cognigy's online resources include the online product documentation and the community discussion thread where issues and projects can be discussed in an open forum. In addition, Cognigy technical support options are outlined in SLAs and customized for each client.

Is it possible to disable analytics collection for specific channels?

Download the full list of responses with over 30 additional answers.

Security & Data Privacy

Question Answer Link

Does the product include features for user management including control for access to virtual agent resources?

Cognigy.AI is enterprise ready with a built-in user management module supporting SSO providers such as Azure Active Directory and Google. The access control module allows role-based and granular user permissions to be created and granted for platform accounts.

What data security capabilities does the product include?

Cognigy.AI complies to industry standards by offering a data security package including environment time-to-live settings, encrypted tokens and chat history storage control options on an infrastructure level.

Please list relevant certifications for privacy, security etc.


Is it possible to manage multiple tenants of the application from a single user interface?

Download the full list of responses with over 30 additional answers.

Implementation & Support

Question Answer Link

What training options do you offer for the product?

Alongside interactive user journeys found within the Cognigy.AI platform, four levels of training courses are provided by Cognigy that start with the fundamentals of conversational AI design and progress through to explore the more advanced features of the platform. Cognigy provides a free demo environment for new users to practice building their projects and the full product documentation is available publicly which explains to a technical level of detail how each feature is used.

Does the product offer support for developent, quality assurance and production deployment pipelines?

Cognigy.AI offers the snapshots feature that allows complete virtual agent projects to be exported and imported to staged environments. In addition to snapshots, resources belonging to the agent NLU such as lexicon libraries and intent training records can also be exported and imported to staged environments.

What deployment options are supported by the product?

Cognigy offers SaaS, dedicating hosting and on-premises deployment options with support for Kubernetes as a container orchestrator. This allows the Cognigy.AI platform to be automatically or manually scaled to match the resource requirement of the customer.

How does your company handle the roll out of platform software upgrades?

Download the full list of responses with over 30 additional answers.