MARK HERSCHBERG. Educated at MIT (with degrees in physics, EE/CS, and a masters in cryptography) Mark has spent his career launching and fixing new ventures at startups, Fortune 500s, and academia. Mark has worked at and consulted to number startups typically taking on roles in general management, operations, and technology. He has been involved from inception and fundraising through growth and sale of the company. These startup companies have included a wireless application platform, online advertising, lead generation, OLAP, media, entertainment, and new language development. Mark was instrumental in launching ServiceLive.com Sears online home services labor market; he also helped fix NBCs online video marketplace (now Hulu.com). In academia Mark spent a year at HBS working with two finance professors to create the upTick system now used to teach finance at many of the top business schools. At MIT Mark helped to start the Undergraduate Practice Opportunities Program at which he teaches annually along with his teaching at the SUNY Levin Institute. In his spare time Mark works with numerous non-profits, including his work on the advisory board of the Museum of Math, and was one of the top ranked ballroom dancers in the country. Mark is currently the CTO of Madison Logic.
Tell me about your firm (number of employees, location, type of companies you work with, etc.).
Madison Logic is the premier provider of intent data solutions for 1,000+ of the world’s leading B2B marketers and publishers. The company’s Content Consumption Monitoring technology provides intelligent lead cultivation and monetization that ensures B2B brands are able to maximize the value of the leads they cultivate from acquisition to loyal customers. Madison Logic’s technology also empowers premium publishers to more efficiently monetize their businesses. With solutions driven by intent data, business buyers receive messages relevant to every stage of their journey to purchase.
Today, more than 600 of the top B2B marketers depend on Madison Logic to maximize the reach, efficiency, effectiveness, engagement and insights delivered by their campaigns. The company maintains long-term partnerships with a highly refined base of 450 premium B2B publishers, who rely on Madison Logic to help them drive audience growth and greater revenue.
Tell us about the challenges facing IT companies today?
Systems, both online and physical, are producing significantly more data every year, hence the rise of “Big Data.” With this typhoon of data companies face a number of challenges: capturing the data, storing it, accessing it efficiently, and making use of it.
What is your firm doing to overcome these challenges?
At Madison Logic we’ll be capturing hundreds of millions of events this month across our network of partners. We are members of the NAI (Network Adverting Initiative), our industry’s organization championing the responsible and transparent use of information for Interest-Based Advertising. We are committed to privacy and review our process with the NAI to insure we have responsible business and data management practices.
As we’ve continued to grow we have watched our data traffic increase exponentially. Prior to starting this product we looked at the type of data we’d be getting, and how it would be used. After careful analysis we picked Mongo, a NoSQL database.* It allowed us to capture and store the data in a flexible way–critical for a new data business which will evolve–while also being able to scale up to many terabytes of storage.
Using both in-house experts and contractors we’ve developed complex models of intent. The models evolve and improve over time and we continue to add more ways for our customers to consume the data. The net result is that now we have tools that make sense of this mountain of data, allowing our customers to easily identify executives and companies interested in their products and services.
[Note to editor: If you want more generalized thoughts on how other companies can think about big data]
The hardest challenge with Big Data is knowing what to do with it. Even once the data is captured– most companies don’t know how to analyze it or use it make decisions. When people talk about Big Data they often think of highly educated data scientists. For advanced modeling this is necessary. Still, for many companies, there’s low hanging fruit. Don’t worry about “Big Data” and instead start with “Some Data”–capture what you can, do basic analysis (done with basic college statistics) and use it to gain insights: who are your most profitable customers and how are they profitable? What percentage of visitors are one-time visitors and where are they coming from? What products or pages are commonly viewed together? One word of caution: make sure whatever person or team is doing the analysis understands both the mathematical techniques and how the business and industry work. Otherwise you’ll end up with those statistics Mark Twain warned you about.
[Note to editor: If your readers need more clarification on what NoSQL databases are]
Traditional SQL databases store data in rows and columns, you can think of the tables of a SQL database like a giant spread sheet. No SQL databases can be more like a text document, or a graph, or any number of less rigid formats. Both SQL and NoSQL have distinct pros and cons. In our case we selected Mongo which is a document database as it allowed us to build appropriate models and the flexibility to easily enhance and adjust the product as customers find new uses for the data. Along with open source technologies like Storm and RabbitMQ we can process tens of millions of events per hour and can scale up and down during peak load times. This ensures that busy periods don’t overwhelm our systems.
Contact information: Provide website so that we can mention your firm on the radio (Business Talk 1110 AM and on our news site).