In the age of big data, many companies believe that they are effectively using their data. To paraphrase Woody Allen, it’s as if 80% of effective data management is just in the collection. We at Relativity6, however, prefer to break down data, and its multitude of uses, in the way Prof. Tom Davenport has described — between offensive data and defensive data.
At Relativity6, we consider ourselves a “data offense company.” We use machine learning and artificial intelligence to help businesses glean customer insights and make sound strategic decisions in resource constrained environments. Our methods focus on parsing customer data to help drive customer re-acquisition for our clients.
One of the most critical steps to any machine learning process is data collection. At Relativity6, data collection consists of receiving data in all shapes and forms. Our system of algorithms accepts data from CRMs, transaction tables, product tables, and even call center records. For a number of reasons, more often than not, data sets are riddled with empty fields, or are drawn from a disparate number of sources that do not easily align for analysis. Once we begin working with our clients, we clean and transform their data, in order to make it consumable for our algorithms to process. In a perfect world, all the data would be uniform and consistent. Unfortunately, we don’t live in a perfect world.
Gaps in data are to be expected in any large, sophisticated organization. However, such gaps can create huge bias problems. This kind of bias can impact the integrity of the entire strategic solution that an entity has gleaned from its data, and as a result it might be better off tossing a coin than working with so many empty fields. So while data collection isn’t the most attractive element of machine learning, it’s certainly one of the most critical. This is why we at Relativity6 have narrowed our focus to play data offense. We know the sort of data we are looking for because we know the kind of solution we’d like to bring about. It is not just the quantity of data that drives sound strategic decisions in the big data age, but the quality of the data being applied to the problem that is in need of a solution.