PatternBuilders brings big data analytics down to size

Terence Craig
Terence Craig, CEO and CTO of PatternBuilders, discusses how big data and privacy are changing business models at the Private Identity Innovation Conference in Silicon Valley.

REDMOND, Wash. — May 23, 2013 — Data technology experts and startup veterans Terence Craig and Mary Ludloff founded PatternBuilders with a singular vision: to make it practical and easy for enterprises everywhere to harness big data and make better, faster decisions — without hiring a team of experts.

“We found it disconcerting that there was such a huge divide between big data excitement and actual adoption rates,” said Craig, CEO and chief technology officer at PatternBuilders. “Taking advantage of big data analytics often requires a budget, toolset and in-house expertise far beyond what most enterprises can muster. Mary and I founded PatternBuilders because we thought there must be a better approach.”

Big data analytics is about finding answers to questions derived from data with high rates of volume, velocity or variety (singularly and in concert with one another). Enterprises can use big data analytics to investigate many types of business problems, such as the following:

  • A publicly traded enterprise wants to assess the impact of external social media activity on its stock price, based on factors including message sentiment, time of day, geographic location, and Klout scores of the people posting or tweeting.
  • A nonprofit health organization needs to quickly and accurately recommend ways to halt the spread of a highly contagious disease outbreak.
  • A municipality wants to use IP-enabled parking meters to simultaneously optimize parking revenues, improve traffic flow, and ensure that parking policies don’t discourage customers from visiting retail stores in the city.

“We’re not the classic young kids in a garage,” said Ludloff, co-founder and vice president of Marketing at PatternBuilders. “We’ve both been involved in data-related startups for many years. With PatternBuilders, we’ve been able not only to apply our expertise gained from past startups but also to build on the ‘Big Data 1.0’ tools that have been developed by others already.”

Enterprise-ready big data analytics

Big data analytics tools can require either specialized in-house software-engineering expertise or long-term, on-premises consultants to write customized code. To make big data analytics tools accessible to a wide range of enterprises, the PatternBuilders founders knew their solutions had to be affordable, usable and customizable by a company’s existing IT staff as well as empower business users to perform big data analysis.

Mary Ludloff
Mary Ludloff, co-founder and vice president of marketing at PatternBuilders, speaks at the O’Reilly Strata Conference in New York City.

“In-house big data analytics tools based on Hadoop or other open-source technologies are fine in environments with sufficient software engineering resources,” said Ludloff. “But only a handful of organizations globally fit that profile.”

Craig and Ludloff used experience they gained in the retail industry, where the demand for high-volume, high-velocity, real-time analytics is enormous and budgets are tight, to create tools that are within reach of the majority of enterprises.

“At first, most enterprises we talked to assumed they could figure out how to do a big data analytics solution on their own, but then they found out it’s really, really difficult,” said Craig. “Making the tools easy to use and scalable requires hiding a lot of very complex multiprocessing and multithreading issues.”

PatternBuilders’ first products include pre-customized solutions, such as its FinancePBI for the financial services industry, and a standard, enterprise-ready analytics application (AnalyticsPBI) that analysts or quants can easily customize to meet all their company’s big data analytics needs.

“We’ve made sure that our customers can take our solutions into their companies and start using them right away,” said Ludloff.

Microsoft technologies through and through

PatternBuilders made a conscious decision to wed itself to Microsoft technologies.

“When we started evaluating technologies for PatternBuilders, we were impressed by Microsoft’s multiprocessing capabilities, the cohesiveness of the .NET Framework, its developer toolkits and the Windows Azure cloud platform,” said Craig.

Craig cites a number of reasons for building PatternBuilders’ solutions on Windows Azure. From a business perspective, Microsoft worked hard to make the Azure platform affordable as PatternBuilders transitioned from on-premises and more traditional infrastructure services to a hybrid solution utilizing both Azure Infrastructure-as-a-Service (IaaS) offerings and rich Platform-as-a-Service (PaaS) features. On the technical side, Craig was impressed with the Windows Azure platform’s predictable I/O performance, ease of deployment, natural integration with Visual Studio and the .NET architecture, and support for both platform services and heterogeneous infrastructure services, which is important for PatternBuilders’ use of 10Gen MongoDB running on Linux virtual machines. Windows Azure supports MongoDB directly through 10gen, as well as through the Windows Azure store via MongoDB partner MongoLab.

“As an independent software developer, I don’t like surprises with my technology,” said Craig. “With Microsoft, we seldom get surprised, thanks to its interconnected vision and strong tooling. Also, the widespread familiarity with Microsoft tools means we can achieve tremendous productivity with fewer senior engineers — leaving our senior folks free to focus on the truly tough problems. Plus, the ubiquity of Windows in the enterprise and Azure features like Virtual Networks make it easy for us to work with our customers’ existing infrastructures.”

For the PatternBuilders founders, entry into the Microsoft BizSpark program strengthened their ties to Microsoft technologists, providing direct access to experts who can answer questions and help solve technical issues quickly. As for access to funding, PatternBuilders has chosen a bootstrapped approach for now, although it will likely seek venture capital at some point in the future.

“We’ve both dealt with raising capital from venture funds a few times already, so we know how to get started and what to expect,” said Craig. “Still, we’d be crazy not to take advantage of our BizSpark relationships when we’re ready to look for venture funding. And beyond any funding events, the Microsoft sales and marketing people we’ve met through BizSpark will be enormously helpful for co-selling our products into the enterprise, where Microsoft’s reach and reputation are unparalleled.”

Related Posts