Reduces engineering time, accelerating time-to-market and enabling operational data ingestion and analytics faster than on other platforms
Simplifies data analysis and streamlines decision-making with BigQuery
Allows non-technical end users to explore data more securely and accurately
CellPoint Digital is partnering with Google Cloud to rebuild its infrastructure for scalability and data-driven solutions. Launching in 2025, the new platform will enhance payment experiences for travelers.
The travel industry is making a strong comeback after the challenges of the pandemic years. However, increasing numbers of people on the road (and in the sky), alongside a significant increase in cashless payments, highlights a need for industry players to optimize their payment systems. Failing to do so could mean losing out on significant revenue opportunities—according to a recent study, airlines could capture up to $14 billion in value through strategically addressing payments.
“Payments in the travel industry are complex,” says Mike Williamson, Data Architect at CellPoint Digital. “There are a lot of different currencies and banks involved, in addition to new marketing frameworks in which multiple services such as flights, hotels, and car rentals are bundled into one unified booking system. While this is an exciting shift, the integrations involved mean significant technical challenges.”
CellPoint Digital’s payment orchestration platform acts as a bridge between banks and payment methods, helping to ensure payments go through smoothly and enabling airlines and other travel retailers optimize their revenue. “A major benefit of our platform is our ‘fall-through success’ feature. If one payment provider fails, we automatically try others the airline is connected with until we find one that works,” Williamson explains, “On average, this process delivers a 20% lift in conversions. Customers are impatient. If their payment doesn’t go through on one site, they’ll likely move to another.”
Founded in 2008, the company increasingly felt that its legacy platform didn’t offer what was needed to support rapid growth. Two years ago, they made the decision to leverage the simplicity and scalability of Google Cloud to rebuild their platform from the ground up. “With our small team, we knew creating another in-house solution wouldn’t be sustainable,” Williamson says, “We handle significant volumes of data, and tools like Pub/Sub and BigQuery save us enormous setup time.”
With our small team, we knew creating another in-house solution wouldn’t be sustainable. We handle significant volumes of data, and tools like Pub/Sub and BigQuery save us enormous setup time.
Mike Williamson
Data Architect, CellPoint Digital
Williamson took responsibility for designing and developing the architecture for the data engineering portion of the new platform, called Insights. This encompasses archiving, reporting dashboards, data governance, and data discovery among other components.
With the support of Premier Managed Services Google Cloud partner Devoteam, Williamson’s team is building a microservices architecture running on Google Kubernetes Engine (GKE) within Google Cloud. “Over the years, our legacy platform became a mix of local systems, cloud components, and some decentralized Kubernetes structures. This made it inefficient for scaling, and ill-equipped to handle sudden traffic spikes,” says Williamson. “One of the main reasons Cellpoint decided to go with Google Cloud was for Google Kubernetes Engine. As a data engineer, I was excited to work with Google Cloud because of its phenomenal data products like BigQuery and Pub/Sub.”
The platform is a three-tier structure, with a first tier integrating with merchants, the second serving as a core layer ensuring consistency across users, and the third connecting to payment providers and fraud-checking agents.
One of the main reasons Cellpoint decided to go with Google Cloud was for Google Kubernetes Engine. As a data engineer, I was excited to work with Google Cloud because of its phenomenal data products like BigQuery and Pub/Sub.
Mike Williamson
Data Architect, CellPoint Digital
BigQuery is amazing. It organizes data in a way that avoids complicated setups and lets me access related information easily and instantly, so I can focus on analyzing the data without dealing with technical hurdles or delays. Pub/Sub is so user-friendly a child could set it up.
Mike Williamson
Data Architect, CellPoint Digital
“Google Cloud has clear advantages for data processing, data access, and migration. From my experience, migrating data is orders of magnitude cheaper than on other platforms,” Williamson says. At Cellpoint, data is divided into two categories: operational data for application use and analytical data for reporting and insights. Real-time data is transmitted from the operational services through Pub/Sub and ingested into analytical services. One service writes raw data directly to BigQuery to ensure nothing is lost, while another sends it to object storage, acting as a backup and addressing potential schema issues with BigQuery ingestion. The raw data is then processed using Dataflow and SQL through DBT, transforming it into structured formats ready for reporting, dashboards, and further analysis.
This structured data is integral to two downstream applications: a managed self-service platform that supports customer service agents with tasks such as refunds or payment issues, and a data exploration and reporting system based on Looker enabling non-technical end users to analyze and visualize data using user-friendly dashboards.
“BigQuery is amazing,” Williamson says, “It organizes data in a way that avoids complicated setups and lets me access related information easily and instantly, so I can focus on analyzing the data without dealing with technical hurdles or delays. Pub/Sub is so user-friendly a child could set it up.”
The team particularly appreciates BigQuery’s granular access control, which enables them to create materialized views with separate access rights. This makes it easy to aggregate data and share it securely with a broader audience, providing meaningful insights without exposing sensitive or proprietary information. In addition, BigQuery’s public datasets allow data to be shared in a way that’s useful for analysis but still protects individuals' privacy and complies with regulations. “Data is incredibly valuable, and features like this help balance utility with privacy and compliance requirements like GDPR and PCI,” Williamson says.
Looking ahead, the company plans to explore machine learning opportunities, particularly optimizing payment service provider routing and improving transaction efficiency. By leveraging a supervised learning model, the platform could dynamically adjust provider priorities in real time and enable CellPoint Digital to offer consulting services to merchants, notably identifying cost-effective providers and flagging underperformers. The same system could enable the company to leverage its data to provide broader market insights linked to air traffic forecasting trends.
With the new platform set to launch in 2025, the company is anticipating many more years of collaboration with Google Cloud. “Google Cloud’s serverless tools drastically reduce infrastructure costs and engineering time,” Williamson says. “They just work. I’m looking forward to seeing where the future will take us.”
Google Cloud’s serverless tools drastically reduce infrastructure costs and engineering time. They just work.
Mike Williamson
Data Architect, CellPoint Digital
CellPoint Digital simplifies payment processing for the air, travel, and hospitality sectors. Connecting 28+ card schemes and 168+ alternative payment methods, it helps businesses optimize payment acceptance, reduce costs, increase revenue, and provide a better customer experience.
Industry: Finance
Location: Denmark
Products: BigQuery, Google Kubernetes Engine (GKE), Dataflow, Looker, Pub/Sub, Cloud Scheduler, Cloud Run, Cloud Spanner, Apigee, Cloud Build, Cloud Deploy
About Google Cloud partner — Devoteam
Devoteam has been a Google Cloud partner for over a decade. Its team of 650+ Google experts with 10 Google Cloud Specializations has helped hundreds of organizations grow with Google Cloud, and moved over 1 million users to Google Workspace.