RAPP Leverages the MapR Distribution on Google Compute Engine to Create Differentiated Customer Experiences

RAPP’s progress todate working with MapR’s Hadoop distribution on Google Compute Engine is enabling the agency to develop differentiated, data-driven customer experiences for their clients around the world. They are able to stay ahead of the curve in the data landscape and develop innovative product offerings while increasing efficiency.

The Business

RAPP is a leading full-service global marketing agency that works with clients in over 80 countries to develop engaging customer experiences that create better relationships between brands and customers. Data and insights about customers are the foundation of RAPP’s approach to solve complex business challenges and a source of their competitive advantage.

RAPP’s Data Science department collects, organizes, and analyzes data to understand a consumer’s preferences and behavior so companies can communicate with each individual in a way that is most meaningful.

The Challenge

As RAPP was striving to create more differentiated client experiences that are data driven, they realized they needed a hosted compute and storage environment for testing and development of new solutions. The RAPP IT team was very efficient in serving the needs of clients and billable work, but according to David Sogn, VP of Data Science for RAPP, the challenge was to find a solution that provided the ability to experiment iteratively and roll out rapidly. For example, when they had a theory, they needed the ability to analyze the data and test it.

MapR Solution

In Sogn’s previous job, he used a lot of open source software, and quickly learned to appreciate the growing availability of cloud-based compute and storage environments where he could rapidly prototype automated analysis, visualizations and other data solution. After some trial and tribulation with the early dominant flavors of cloud infrastructure, like AWS, Azure, Rackspace and others, Sogn was won over by the Google Cloud Platform’s simplicity, power and partnerships.

“Partnerships like MapR’s distribution for Apache Hadoop, was like icing on the cake. In the past, my team was distracted from their primary objective’s as data scientists by lost jobs or other issues related to distributed metadata. I was already familiar with the MapR from conferences and meet-up groups, so it appeared to be a great match from the beginning. The MapR people even helped my team better utilize some of the individual component features available on Google Compute Engine that we might have otherwise overlooked.”

The Google Compute Engine and MapR were key to RAPP’s ability to get up and running quickly; and now they’re fundamental to nearly all of RAPP’s data-intensive projects.


The combination of MapR on Google Compute Engine enables the RAPP team to focus on improving their efficiency and developing new products which grow their business.

Delivers on reliability and creates peace of mind
“Google Compute Engine and MapR are key to our team being successful. We view our data as a corporate asset, and now we’re more confident our data is safe,” stated Sogn. “MapR software does what it says it will do — that’s really valuable. When you install a piece of software and have it run as described in the documentation and fix itself, it is not only simple, it’s smart. Now factor in Google Compute Engine’s massively scalable infrastructure and sub-hour billing feature and it’s a slam dunk”.

Increases efficiency
RAPP sees the Google Cloud Platform as key benefit by providing additional tools to tackle problems more efficiently and improving responsiveness to client needs. For example, the benefits of open source plus integrated innovations enables RAPP to benefit from the most recent releases of an open source project with verification and full support from MapR.
“Eliminating tasks unrelated to analyzing, building models or interpreting data represents a significant productivity boost for team members,” states Sogn. “Whenever I can have my team working in the development environment, analyzing data instead of doing load balancing or running inventory checks, that increases our efficiency. This efficiency can then be passed on to our clients.”

Enables new product offerings
RAPP is in the process of releasing a new client solution on Google Compute Engine — a messaging optimization console that helps companies determine with whom to communicate, what channel to use, and what message or media to deliver to that individual.

“There is a lot of information you need to bake into a solution like this when making cross-channel campaign recommendation,” says Sogn. “You’re dealing with millions of IDs for one client. Google Compute Engine is able to store and process massive amounts of data quickly and MapR’s speed is critical to our solution offering.”

A leader in the changing data landscape
RAPP feels confident moving forward with Google and MapR as their partner — particularly with their ability to combine the best of open source with innovations that provide ease of use, dependability and performance advantages. “One of the reasons why it’s smart for us to stick with this solution is we’re confident that they’re leading the way. For instance, MapR appears to be plugged into the ecosystem and is therefore able to strategically help companies better use the information they have,” said Sogn.

mapr_case_study_rapp_4.pdf879.89 KB