When we describe ChatGenie, we don’t start with “chatbot” or “ticketing system.”
We start with Agentic AI Systems.

ChatGenie builds multi-agent AI systems that autonomously analyze, decide, and act across complex enterprise workflows. Customer support operations are simply the first — and most visible — foundation where this approach creates tangible impact.

This case study highlights how a leading motorcycle ride-hailing platform in the Philippines transformed its customer support operations using ChatGenie AI Agents, achieving:

• 77% reduction in support operations OPEX
  
  
• 10x faster time-to-resolution (from 5 minutes to 30 seconds)
  
  
• Support headcount reduced from 39 agents to 9 agents across 3 shifts
  
  
• All while handling the same ticket volume, at a scale of hundreds of thousands of daily bookings and tens of thousands of riders

‍

The Challenge: High-Volume, High-Complexity Support

The client operates one of the largest motorcycle ride-hailing networks in the country, with:

• 1M+ social followers
• ~250,000 average daily bookings
• 50,000+ active riders
  
  

Support volume is high and highly contextual. Customers and riders message through social channels with concerns ranging from booking issues and payment problems, to rider onboarding and account verification.

Before ChatGenie:

• 39 human agents were spread across three shifts
• Response and resolution times varied widely
• Complex or sensitive cases weren’t always routed to the right person quickly
• Support leaders lacked real-time visibility into sentiment and service quality

The team needed more than another chatbot. They needed an AI system that could understand users, triage and prioritize issues, and collaborate with humans, not just auto-reply to FAQs.

‍

Our Approach: Agentic AI, Not Just Automation

Instead of deploying a single monolithic chatbot, ChatGenie implemented a multi-agent AI framework purpose-built for this client’s operations.

We designed and orchestrated several specialized AI Agents working together:

1. Routing & Risk AI Agents

Human Escalation AI Agent

• Purpose: Detects conversations that are high-risk, complex, or sensitive and routes them to human agents.
• What it does: Scans messages for signals like strong negative sentiment, regulatory keywords, dispute language, or repeated failed attempts at resolution. When triggered, it flags the conversation, assigns the right queue or specialist team, and provides a quick summary so the human agent can jump in with context.

‍
Auto Categorization & Priority Setting AI Agent

• Purpose: Organizes and prioritizes every incoming conversation so the team always works on the right issue at the right time.
• What it does: Automatically tags conversations by topic (e.g., booking issues, payments, rider onboarding, account access) and assigns urgency levels based on content (e.g., active trip vs. general inquiry). This powers smarter queues, reporting, and SLA management.
  
  

2. Understanding-the-User AI Agents

User Identifier AI Agent

• Purpose: Instantly identifies what type of user is messaging (for example, rider vs. passenger) and applies the correct workflows.
• What it does: Uses conversation history, language patterns, and metadata to infer the user role. Once identified, it selects the right reply templates, rules, and routing logic—removing the need to ask users repetitive “Are you a rider or a customer?” type questions.
  ‍

‍CSAT AI Agent

• Purpose: Measures customer satisfaction and detects emerging issues in real time.
• What it does: Analyzes language, tone, and explicit feedback within the conversation. It estimates satisfaction scores, highlights dissatisfaction, and surfaces themes (e.g., repeated complaints about a step in the app) so operations teams can act quickly.
  
  

3. Context Enrichment AI Agent

Image Recognition AI Agent

• Purpose: Converts unstructured screenshots and images into structured data that the system and human agents can use immediately.
• What it does: When users send screenshots (for example, of booking details or error messages), this agent extracts relevant information such as booking IDs, reference numbers, timestamps, and error codes. It attaches these as context to the ticket so the AI and human agents no longer need to ask users to manually type long IDs or describe what’s on the screen.
  
  

Together, these AI Agents form a coordinated system that doesn’t just respond — it observes, decides, routes, escalates, and learns.

‍

Fast Execution: From Intro to Launch in 80 Days

Despite the complexity, the end-to-end rollout was fast:

1. Intro & Alignment – Defined goals (efficiency, speed, and quality) and mapped support workflows
   
   
2. Knowledgebase & FAQ Upload – Centralized key policies, flows, and standard replies for AI grounding
   
   
3. Evaluation & Tuning – Ran controlled tests, refined prompts, agent behaviors, and escalation rules
   
   
4. Due Diligence – Completed reviews on security, compliance, and operational readiness
   
   
5. Go-Live – Launched across production support channels
   
   

Total time from intro meeting to launch: ~80 days.

‍

The Results: Operational Efficiency, Proven in Numbers

After the deployment of ChatGenie’s multi-agent system:

• Support headcount dropped from 39 agents to 9 agents across 3 shifts
  
  
• The client achieved a 77% reduction in support operations OPEX
  
  
• Time-to-resolution improved 10x – from around 5 minutes to roughly 30 seconds for common, repeatable concerns
  
  
• Human agents now focus on complex, high-touch interactions, while AI Agents autonomously handle routine inquiries and triage
  
  

Instead of scaling support by adding more people, the client scaled by adding more intelligence.

‍

Why This Works: Agentic AI as a Systems Layer

Most “AI customer support tools” are built as add-ons — chatbots, ticket widgets, or help center search.

ChatGenie takes a different approach:

• We design AI Agents as first-class system components, not features
  
  
• These Agents plug into existing tools and workflows (inboxes, CRMs, helpdesks)
  
  
• They collaborate with human teams, rather than simply replacing steps
  
  
• We treat customer support operations as the starting layer for broader enterprise AI automation
  
  

In this deployment, the visible chatbot experience is just the surface. The real transformation happens behind the scenes: in routing, prioritization, risk handling, user understanding, and context enrichment — all powered by AI Agents.

‍

What This Means for Other Enterprises

If your organization handles large volumes of conversations across channels — and you’re asking:

• “How do we reduce our support OPEX without hurting quality?”
  
  
• “How can we give our best agents more time for high-value cases?”
  
  
• “How do we get real-time visibility into user sentiment and operational risk?”
  
  

— then you don’t just need a chatbot. You need an agentic AI system built around your operations.

ChatGenie can:

• Map your existing workflows and pain points
  
  
• Design a customized set of AI Agents (routing, escalation, enrichment, evaluation, and more)
  
  
• Integrate with your current tools and data sources
  
  
• Launch in weeks, not years — with measurable impact on cost, speed, and customer experience
  
  

See What Agentic AI Could Do for Your Team

This case study is just one example of what’s possible when enterprises treat AI not as a plugin, but as a systems layer for their operations.

If you’d like to explore how a multi-agent AI framework could work in your organization, reach out to us at ragde@chatgenie.ph.

‍