About the client
The client is a small company in the travel industry developing a digital product to manage travel assistance and requests. The product addresses the needs of small and medium-sized businesses, where travel coordination often involves multiple communication channels, travel services, and manual interactions between customers and travel managers.
Business context
Travel coordination for small and medium-sized companies often happens across disconnected tools. Customers communicate via email and messaging apps, while using separate services to search for flights and other travel options.
This fragmented workflow makes it easy to lose context, duplicate information, and makes the overall request difficult to track. More complex trips create additional operational pressure because multi-service and non-standard requests require coordination across several touchpoints.
Many existing products focus primarily on self-service. They offer fewer options for travel managers who need to oversee customer requests, communication, and service operations within a single environment.
The client set out to create a product that could reduce this fragmentation and provide a more structured way to support travelers. The MVP focused on 2 core components: an AI-powered concierge for customer assistance and an administrative platform for managing the service it supports.
We had a really good working rhythm with the client. We stayed in close contact, talked through priorities together, and adjusted the plan when needed. That kind of collaboration made it easier to keep the MVP focused and make sensible decisions for the next stages.
Ivanna Avksentieva
Senior Project Manager
Fragmented travel request management
Customers relied on email, messaging, and separate travel services rather than a single environment to create and manage requests.
Manual coordination across multiple channels
Disconnected communication increased the risk of lost context, duplicated information, and harder process control.
Complex multi-service requests
Travel managers needed a more manageable way to handle requests involving several services or non-standard requirements.
Limited operational visibility
Existing workflows made it difficult to maintain a clear view of request status and interaction history between customers and travel managers.
Solutions we implemented
We started the engagement with a structured discovery phase to understand the client’s business processes, validate requirements, and determine which capabilities to include in the first release.
Together with the client, we divided development into prioritized phases rather than committing the full product scope upfront. The project followed Agile Scrum with 2-week sprints, regular client workshops, sprint reviews, and continuous alignment of requirements.
AI-powered concierge chatbot
We developed an AI-powered chatbot that acts as a digital concierge for travelers.
The assistant provides users with quick access to travel-related information and support without requiring direct involvement from service personnel in every interaction. OpenAI API provides the AI capabilities behind the concierge.
Centralized administrative platform
We implemented an admin platform that provides the client’s team with a single environment to manage and monitor the concierge service.
Administrators can manage chatbot content and responses, work with users and system data, oversee chatbot operations, and maintain service configuration from the same interface.
Discovery-led product roadmap
Rather than moving directly into full-scale development, we worked with the client to validate business processes and prioritize the functionality that would deliver the highest value for the MVP.
This approach helped define a clear product roadmap and kept the initial scope focused on the capabilities needed to test the product concept with lower implementation risk.
Phased product delivery
In collaboration with the client, we divided functionality into several prioritized delivery phases.
This allowed the core capabilities to reach MVP earlier, spread investment across project stages, and preserve flexibility to adjust later phases based on product feedback.
Third-party integration roadmap
The team identified the external integrations required by the product and planned them across individual development stages.
Only the integrations needed for each phase were included in the immediate scope, reducing initial technical complexity and keeping the MVP focused on its core functionality.
Service configuration
Centralizes the setup and management of the concierge platform, including core service settings and operational parameters.
AI-powered travel concierge and monitoring
Provides travelers with AI-assisted travel support while giving administrators visibility into chatbot activity and interactions.
Chatbot content management
Allows administrators to manage the content and responses that guide the concierge experience.
User administration
Provides centralized control over platform users and related system data.orted channels
The project has recently reached the MVP stage, so adoption, operational efficiency, and customer satisfaction metrics are not yet available. Current results focus on product delivery and validation of the core concept.
MVP was released with the core product capabilities
The first release brought together the AI-powered concierge and administrative management functionality required to validate the product concept.
This gives the client a working foundation for testing the service before committing investment to the broader product roadmap.
Core functionality reached users earlier through phased delivery
Splitting the product into prioritized phases allowed the team to focus the first release on the highest-value capabilities instead of waiting for the full planned scope. The approach also gives the client room to adjust subsequent phases based on feedback and product needs.
Product scope remained flexible for future integrations
The team mapped third-party integrations across individual project phases and limited the MVP to integrations required for the initial product scope. This reduced complexity in the first release while preserving a defined path for adding further capabilities.
Client expectations met at the MVP stage
ESIGN Act and eIDAS compliance, backed by a complete audit trail.
Business impact measurement comes next
Because the MVP has only recently been released, the project does not yet have validated metrics for operational efficiency, user adoption, revenue, or customer satisfaction.
These results should be added once sufficient usage data becomes available.
