There is an undeniable global movement taking hold: the growth of data-driven decision making. Forbes contributor Dileep Rao recently commented on this phenomenon in his article, “Why Old-School Bankers Will Become New Age Dinosaurs.” In the last few decades, the vast power of modern technology has allowed for unprecedented volumes of information to be generated and collected. Now, with all this data at hand, both existing and new companies are racing to find the best way to manage, analyze and utilize all this material. The key is no longer just gathering data, but transforming that information into a valuable asset.
Already we have seen short-term, operational decisions dictated and carried out by data, with little to no human intervention. Automatic decisions are made by computers analyzing past and current data. For example, a store’s inventory database notifies the manager when an item is sold out or GPS software gives directions based on maps and traffic analytics. Strategic decisions, on the other hand, are largely still made by people instead of machines. These choices generally involve a combination of long-term policies and data analysis with a far less predictable outcome. The necessary data is accumulated, mined, and interpreted by lower-level analysts, presented up the ladder, and then the decisions are made by top-level management with this compiled information at hand.
The new wave of data-driven businesses is moving to minimize the necessity of human analysis and choice, which takes up the time and effort of employees at additional cost. Instead software is now able to create far more sophisticated analysis, and furthermore, predict future values and choices. As a result, decision-making is divided into two camps. The first involves analytics that present a clear cause and effect, where software can use sophisticated algorithms to make immediate decisions behind the scenes. The second camp does not present an immediate correlation, and as a result requires human reasoning. Even so, the software still allows for far more informed choices based on the vast amount of accumulated information.
This movement is undoubtedly affecting the lending market in a big way. The recent influx of capital into the small business lending arena by alternative lenders, such as peer to peer and merchant cash advance lenders, has created high demand for loan product that can be quickly evaluated and disbursed. Borrower and deal eligibility are based upon historical data analytics, providing competition for “old school” banks and their methods. The practices traditionally used to determine credit-worthiness – cash flow analysis, collateral, credit history, equity, etc. – are time-consuming and often subjective.
It has often been said that small business lending is an art and not a science, but perhaps it is a little bit of both. The crash of the financial markets in 2008 and 2009 was preceded by a period of robust lending activity to small businesses and home buyers. At the time, lenders were using not only the “old school” methods but in many cases (particularly as competition increased), also quick decisions utilizing credit scoring. However, when a financial crisis hits, all aspects of the market are rocked: cash flow disappears, collateral is devalued, and equity evaporates. Credit scores also plunge as financial pressure mounts, and not necessarily as a reflection on the individual. Thus neither traditional methods nor data-driven decision making have proven to be predictive, although they both may work during good times.
The U.S. Small Business Administration has been using loan scoring for many years. Due to accuracy in historical data analytics, this year the administration announced that for all SBA guaranteed loans of $350,000 and under, participating lenders must use the FICO SBSS (Small Business Scoring Service) score. If a loan score exceeds the mandated minimum, the lender need not analyze the borrower’s cash flow and ability to repay. This system undoubtedly adds great efficiency and standardization to the process, providing lenders with the ability to quickly process loans. It also further enables entrepreneurs to gain capital for their small businesses, many of which would otherwise go without it or be forced to obtain it at higher rates from alternative lenders.
As valuable as loan scoring is, we must recognize that scoring models are based upon an analysis of vast pools of loan data. Although they may provide a probability of how a group of similarly scored loans may perform, they cannot predict the outcome of an individual loan. That is where the banking maxim of “know your borrower” comes into play, combining art and science to narrow the risk.
For small business owners seeking loans, this process can be particularly frustrating when it is unclear what lenders are looking for or what their score standards may be. Instead of going bank to bank and being met with endless paperwork, technology such as bQual.com gives borrowers the ability to instantly get their SBSS small business score and understand their loan fundability. It also gives lenders a data-driven, standardized way to pre-screen borrowers and potential deals. For both borrowers and lenders, this automated process and access to key data eliminates the need for intermediation. The BoeFly.com online marketplace provides even more efficiency by breaking down geographic barriers, allowing for secure and immediate loan connections on a national scale.
Data analysis is no longer a simple commodity for other companies to use at their disposal, the technology itself is driving an entire sector. Advanced software is able to gather increasingly complex analytics, and those able to transform that analysis into accessible information are just as valuable to its utility. These innovations are quite simply easier, faster, and more accurate than their predecessors. Businesses must learn to adapt to this modern, data-driven world or else be left behind.
Rao, Dileep. “Why Old-School Bankers Will Become New Age Dinosaurs.” Forbes.com (July 8, 2014).