The Level Group (TLG), a global leader in luxury goods distribution and logistics, manages customer support operations for some of the world’s top fashion and lifestyle brands. As TLG expanded internationally, they faced critical challenges: customer support ticket volumes grew exponentially, agents were overwhelmed, and response times suffered. Their existing workflows were fragmented, requiring support agents to jump between multiple platforms like Zendesk, Kibo, Qapla, and internal databases just to retrieve basic information.

These operational bottlenecks made it increasingly difficult to maintain TLG’s high standards of service. Agents spent too much time on manual data gathering and categorization instead of resolving customer issues. Time-to-close (TTC) metrics were too high, customer satisfaction was at risk, and scaling the support team through hiring was unsustainable.

Rather than immediately building a solution, Stackdrop started with a structured discovery phase. We worked closely with TLG’s operations managers, support agents, and IT teams to map the full customer support workflow, identify bottlenecks, and deeply understand the technical and operational pain points.

Workshops and Discovery Calls: To uncover hidden inefficiencies, document key workflows, and define critical metrics.

Data Mapping: We studied how information was siloed across tools and identified the most frequent data needs during ticket resolution.

MVP Definition: We designed a Minimum Viable Product (MVP) that prioritized impact, centralizing agent workflows while laying the foundation for future AI automation.

Phase 1: Centralized Support Dashboard Using Retool, Stackdrop built a custom internal dashboard that unified all essential customer data into one clean, easy-to-use interface. Instead of toggling between six platforms, agents now had everything they needed in one place. Features included:

Real-time integrations with Zendesk, Kibo, Qapla, and internal systems.

Instant retrieval of customer, order, and return information.

Unified ticket view and action management for faster resolution.

This centralized system alone transformed agent efficiency and significantly reduced the average time needed to resolve tickets.

Phase 2: AI-Powered Automation With a centralized foundation in place, we introduced AI-driven ticket triage using GPT models customized for TLG’s needs:

Automated Categorization: AI accurately classified over 80% of incoming tickets based on their content.

Suggested Replies: AI-generated response templates allowed agents to approve standard replies with a single click.

Multilingual Capabilities: AI adapted responses based on the customer’s preferred language, enhancing TLG’s global service.

Human Review Safeguards: Complex or unclear cases were flagged for human handling, maintaining quality and control.

Through continuous training and iteration, the AI improved over time, reducing backlog and freeing agents to focus on high-value, complex support tasks.

20,000+ support tickets processed through the new system within months of launch.

Time-to-close (TTC) reduced by over 60%, improving both agent productivity and customer satisfaction.

80% of ticket categorization automated, massively reducing repetitive manual tasks.

Consistent multilingual support, enabling TLG to serve a global customer base faster and more effectively.

Future-Proof Infrastructure: TLG is now positioned to continue scaling operations without needing proportional increases in headcount.

This project not only solved TLG’s immediate operational challenges but also laid the foundation for future AI-powered enhancements across customer support, sales, and service operations.

Retool

OpenAI (Custom GPT Model Integrations)

Zendesk, Kibo, Qapla APIs

Custom Internal APIs and Database Integrations

By combining low-code development, deep operational discovery, and targeted AI automation, Stackdrop delivered a solution that reshaped TLG’s customer support, moving from fragmented manual processes to a centralized, intelligent, and scalable system. This project exemplifies how modern technology can not only solve operational problems but unlock long-term strategic advantage.