Customerly Aura isn’t just another AI assistant.
It’s a fully multilingual, multimodal, and data-aware automation system that thinks, acts, and learns like a true teammate, all while maintaining your brand’s tone, context, and formality across every interaction.
Check the full diagram here on how Aura works.
1. Built for Understanding From the First Message
Every time a new message arrives, Aura begins by analyzing its content and emotional tone.
If the user sounds frustrated or asks directly to speak to a human, Aura instantly escalates the conversation to your team.
Otherwise, it starts reasoning through the message.
At its core, Aura uses a multi-layered understanding system made of:
Intent Recognition: to understand what the user wants.
Missions: to take structured, goal-oriented actions.
Knowledge Base Search (RAG): to find accurate answers.
Confidence Scoring & Escalation: to ensure quality and precision.
The result? Fast, accurate, and human-sounding conversations that adapt to every customer situation.
2. Intent Recognition: Understanding the “Why” Behind Every Question
Aura continuously checks whether a message matches any intents defined inside your Customerly Chatflows.
Intents act as the foundation of understanding for example, “change my plan,” “reset password,” or “check delivery status.
Once an intent is detected, Aura instantly activates the right conversational flow, giving every interaction contextual awareness from the start.
3. Missions: From Conversation to Action
When Aura identifies an intent, it enters Mission Mode: a focused state where it stays until the goal is completed.
Inside a mission, Aura can:
Access any contact or company property stored in Customerly.
Call external or internal APIs to fetch or update data.
Ask clarifying questions, collect inputs, or confirm details.
Reason and act autonomously while keeping the conversation natural.
Each mission has a clear goal, whether that’s updating an account, checking usage, or solving a billing issue, and Aura remains in mission mode until that goal is achieved.
4. Using Data as Context: APIs, Users, and Company Information
Aura doesn’t rely on general AI knowledge.
Instead, it operates entirely within your ecosystem securely using only your verified data.
It automatically fetches context from:
User and company properties (like plan, MRR, language, or region).
Authorized APIs, to retrieve or update real-time data when necessary.
This ensures every response is accurate, compliant, and personalized — no external hallucination, no generic answers.
5. Knowledge-Based Reasoning and Confidence Control
If no intent is found, Aura turns to your Knowledge Base using Retrieval-Augmented Generation (RAG).
It splits complex questions into smaller ones, searches across articles and documents, retrieves the most relevant snippets, and builds a unified, coherent answer.
Each answer is then scored for confidence:
If the confidence score is above 90%, the message is shared with the customer.
If it’s lower, Aura escalates it automatically to a human teammate.
This system ensures your customers only see accurate, verified answers, maintaining trust and satisfaction.
6. Multimodal Intelligence: Understanding Images and Documents
Aura is multimodal by design.
It can analyze images, screenshots, PDFs, or documents shared within a conversation to extract meaningful data and provide contextual guidance.
For example:
If a user uploads a screenshot of an error message, Aura identifies it and suggests the exact troubleshooting steps.
If a PDF invoice or document is attached, Aura reads it, interprets its content, and provides guidance or confirmation based on that information.
This capability allows Aura to handle complex customer support cases that go beyond text — giving your team a powerful AI that sees, reads, and understands like a human.
7. Sentiment Awareness and Human Escalation
Throughout every interaction, Aura constantly monitors sentiment and human requests in real time.
If a user expresses frustration or asks to speak to a person, Aura immediately hands off the conversation, ensuring a seamless transition without customer friction.
8. Smart Learning After Each Conversation
Once a conversation is closed, Aura doesn’t stop.
It performs an automatic topic analysis to tag the conversation with the most relevant categories.
If no suitable topic exists, Aura suggests new ones for your team, helping improve your analytics, reporting, and automation coverage over time.
9. Multilingual by Design
Aura can understand and reply in any language instantly adapting to the user’s preferred language, tone, and level of formality.
At a higher level, every project in Customerly defines its own tone and communication style, which Aura uses as priming context.
This ensures that every response, in every language, always sounds like your brand.
10. Why Aura Feels Different
Aura isn’t a chatbot that replies to FAQs. Aura transforms customer service from a reactive task into a proactive, intelligent experience.
It ensures that:
Every conversation starts with clear intent and context.
Every answer is accurate, personalized, and brand-consistent.
Every escalation happens at the right moment.
Every message — text, image, or document — is understood.
That’s how Customerly Aura delivers impeccable customer service at scale — combining human empathy, AI precision, and your company’s unique voice.