Boxever is a customer intelligence and personalized marketing platform for the travel industry. The Boxever solution enables customers like airlines and online travel agencies to capture, analyze, and act on large volumes of data about customer behavior in real time to create stronger, more profitable customer relationships. The company’s custom analytics engine delivers insights into how each traveler searches, shops, and books travel by synthesizing customer data from multiple, often siloed repositories (such as booking systems, web analytics, mobile activity, loyalty systems and even third-party sources such as social profiles). Armed with these insights, its personalization platform allows travel companies to segment and target each customer individually in real time, and deliver tailored, personalized marketing messages and offers across any communication channel. Boxever is based in Dublin, Ireland. The company was founded in 2011.
The travel industry has become increasingly competitive as potential travelers have access to more information across a variety of sites, apps, and devices than ever before. To help its clients drive the greatest revenue per transaction, increase customer lifetime value, and improve loyalty, Boxever analyzes consumer behavior and provides insights on the fly so that the company’s clients can provide compelling, timely deals to consumers. For Boxever to be successful, it must deliver those insights in real time, while ensuring robust security of consumer behavior data. “Using the cloud meant we would be more agile and more flexible than our competitors,” says CTO Alan Giles. “We needed a solution that would be cost-effective and flexible, ensuring that we could turn on a dime if a market opportunity arose.” The company also needed a solution that would provide low latency and high availability to a worldwide audience.
Why Amazon Web Services
Boxever uses Amazon Web Services (AWS) to process and store customer data. The company captures data in real time and puts it through a data pipeline for analysis. As data comes through its pipeline, it is processed in a Cassandra database and then run through Boxever’s custom business intelligence (BI) and analytics platform. “We capture and process data in real time, so being able to launch instances on demand is very important around sales spikes for airlines,” Giles says. “We are not reinventing the wheel by managing hardware, which allows us to spend our time building innovative solutions and bringing real ROI to our customers.”
The company also uses Amazon Elastic Block Store (Amazon EBS) for backups of data sets and Amazon ElastiCache for in-memory caching. Long-term backups use Amazon Simple Storage Service (Amazon S3). Amazon CloudWatch is used for monitoring the health of Amazon EC2 instances.
Security was a key consideration when Boxever was deciding on a cloud provider. Boxever encrypts all in-flight and at-rest data, leveraging AWS best practices for security to help maintain the confidentiality and privacy of data. “We’re very conscientious about security,” Giles says. “We can comply with security regulations worldwide, in part because of the best practices from AWS.”
Flexibility and speed to market were crucial to the startup. “In the days before AWS, getting capabilities live for research and development (R&D) purposes was a long and tedious process,” Giles says. From initial hardware selection and securing a purchase order, to waiting for hardware to arrive, configuring it, launching the R&D test could take Boxever weeks, if not months. “Couple that with the up-front capital costs associated with purchasing hardware, and the scenario becomes very prohibitive—not to mention risky—for a start-up who wants to move fast and fail early,” Giles says. “With AWS, I could sign up with my personal credit card (which we did in the early days), and within days know if our design and assumptions were valid. The time and cost savings of this approach are nearly incalculable, but are definitely significant in terms of time to market, resourcing, and cash flow.”
Boxever uses Multiple Availability Zones to provide always-on, low-latency service to clients around the world—including Europe, Australia and Japan, where the company is expanding. “We are a real-time operation, and moving our data closer to where our customers are reduces latency phenomenally,” Giles says. “We strive to keep our latency under 100 milliseconds, and AWS helps us accomplish this.”
Using AWS enabled Boxever to launch its business without having to make large investments in hardware. “Using AWS allowed us to adopt a “fail early, fail fast” prototyping strategy by scaling several instances of different sizes and then deciding which instance was most suitable for the job,” Giles says.
The company has decreased both cost and time to market by taking advantage of AWS on-demand pricing. “Using Amazon EC2 Reserved Instances has turned out to be very cost-effective, and that’s a huge selling point,” Giles says. “It’s very easy to scale, so as we get more customer data, we can just increase the number of Reserved Instances we’re using to adapt to demand. That reduces both capital and operational expenditure. For example, we use ElastiCache without any management overhead.”
Using AWS enables the company to spin up instances quickly to test a new business case or adjust to a spike, as well. “I can launch 100 servers if I need to, anytime I need to,” Giles says. “The flexibility that AWS gives us is critical.”
Finally, using AWS enables Boxever to focus on providing real-time insights to its clients, rather than on maintaining infrastructure. “We compete with some very large companies with a lot of resources—but we’re more agile than they are, because we’re not tied to an on-premises data center,” Giles says. “Using AWS gives us a significant competitive advantage—we can be innovative and disruptive, which sets us apart from competitors.”