The famous tech startup bootcamp, Y Combinator, is best known for connecting promising upstart tech businesses with venture capital. Its second and lesser-known but equally important function, is to coach the owners of those startups toward the best possible business model for their organization. What that model turns out to be is determined by many factors like the character, drive, and interests of the people involved, their range of expertise, and their projected potential to service some void in the market.
The famous tech startup bootcamp, Y Combinator, is best known for connecting promising upstart tech businesses with venture capital. Its second and lesser-known but equally important function, is to coach the owners of those startups toward the best possible business model for their organization. What that model turns out to be is determined by many factors like the character, drive, and interests of the people involved, their range of expertise, and their projected potential to service some void in the market. The final conclusion is all based on data, market research, and industry-insider information that Y Combinator, through its influence in Silicon Valley, makes available to its participants. In fact, a high percentage of the 20 or so startups that are invited to participate in Y combinator’s boot camp each year emerge with a totally different product offering than what they originally presented.
Wouldn’t it be nice to have an optimized business model before going to market? Well, because of big data and the analytics tools on the market that have been created to make use of it, these capabilities are now available to any business that wants them.
How can a few spreadsheets do all that?
They can’t. To be of any value, raw data requires an app to curate, organize, and manage it. That app needs a user-friendly dashboard to put data into actionable use. And there is a growing number of businesses that are offering cost-effective, easy-to-use products that bring big data management down to the average user level. There is a growing number of these apps hitting the market. Some are very specialized like hiQ, SumAll, and Splunk. Others like Domo and Hadoop are more general-purpose, turnkey apps for large enterprise. Big data and big data analytics tools are showing businesses how to restructure their model in some key areas. Let’s talk about a couple of examples
The Japanese insurance company, Tokio Marine Holdings, found that some of their customers wanted insurance for policies that covered very short periods, rather than the usual year-by-year model. At the time, their existing sales methods had no answer for that. Conventional wisdom told them to sell long term policies because of the high up-front cost of selling. Short-term policies would never recoup those costs.
To address the needs of their customers, they partnered with mobile carrier Docomo to create a series of short-term insurance products, which they labeled One-Time Insurance. These products were available through a mobile app that provides users with targeted recommendations for certain lifestyle insurance products such as skiing, golf, and travel-related insurance. Through the medium of the app, Tokio was able to create customized, on-the-spot insurance packages.
Capitalizing digital assets
More and more companies are realizing that their data can be turned into a product. A Wall Street Journal report from June, 2014 explains how companies like Facebook and LinkedIn turned search algorithms, housing price predictions, and similar offerings into products that made their sites more popular and successful. But, it isn’t just the strictly digital business entities that have done this. General Electric, Apple, and the New York Times have done it too. Data products don’t have to be stand-alone, or profitable on their own either. In the case of Facebook, data products are often bundled with other offerings or used to draw in viewers to Facebook ads.
When data reveals the limitations of data
One of the most interesting and increasingly common findings drawn from big data is in fact the inherent limitations to the quantitative data model itself. To be more specific, big data tends to yield value because of the quantity of data in the data set. But the USGS has learned that not all data is created equal. With the intent of using social data to develop faster response time to global geological events, USGS refined their social data tracking to search for certain nuances in the language of Tweets. They had noticed how people sharing links or facts related to earthquakes, like magnitude, were less likely to be giving firsthand reports. What they did was restructure their model to filter Tweets sharing a link or a number. What they have now is an earthquake tracking model that uses social data to generate email alerts in as little as two minutes. What’s interesting is that it took a sea of quantitative data to make the qualitative nuances stand out.
Because of insights that can be gained from big data, certain elements of business intelligence there were once only slightly more reliable than prognostication and soothe-saying can now be accurately measured, charted, tracked, and projected into future timelines along a basic if–then logical model. At face value, it’s obvious that big data is a tool that optimizes business processes and reduces inefficiencies. Slightly deeper inspection shows us that big data can do far more than just optimize processes and overcome big data challenges within an isolated information bubble like a single business entity. It can show businesses entities how to optimize their entire model to better fit the market as a whole.