In a world of exponentially growing levels of data where organizations need more insights, both in terms of volume and depth, advanced analytics is at the forefront of driving value. Such analytics reveals hidden patterns and linkages, informs predictions, and discovers opportunities. The more advanced and expansive are the analytic methodologies and capabilities, the greater the potential utility of the results.
At the other end of the sophistication spectrum, basic analytics has more limited value in more limited domains, for example, when scoring customers, fueling dashboards, and defining and tracking toward goals. The general understanding within our industry is that advanced analytics is better than basic analytics, as long as advanced analytics can be generated at scale and in a timely manner.
At the same time, however, for those uninitiated in the inner workings of analytics there is growing concern with black boxes. When computations and algorithms are understood by only a select few, when the processes to generate information and insight are not obvious to the people consuming the information, and when the analytics is complex and not intuitive, its black box nature reduces the value of advanced analytics.
More advanced methodologies, such as machine learning, artificial intelligence, and econometric modeling are considered black boxes by most business decision-makers, especially as this class of analytics continues to grow in advanced capability. Ironically, therefore, the value proposition of advanced analytics is being reduced by its inherent black-box nature. If the primary criterion of analytic adoption is the degree to which it is a black box, then basic analytics will prevail over advanced analytics in informing business decisions. If instead the primary criterion of analytic adoption is sheer analytic insight, accuracy, and depth, then advanced analytics will prevail as a driver of business decision making.
The problem that analysts and decision-makers have is the competing objectives of understandability by the masses versus utility of information to the masses. To resolve this, organizations are increasing investments in analytics departments, and are creating positions such as Chief Analytic Officers, VPs of Analytics, and Lead Scientists, who can vet analytics. By investing in people with these skills who do not consider advanced analytics a black box, organizations can benefit from the value on offer and entrust the black boxes to them.
For organizations that do not have advanced analytics skills, it is critical that third-party suppliers of analytics services change their engagement models from being a vendor to being an analytic business partner. By acting as a supplier to a customer, the customer is responsible for understanding its needs, determining how to leverage analytics, and ensuring organizational adoption of the analyses. The challenge of black boxes, however, inhibits the success of analytics within these organizations.
When third-party analytic organizations act as business partners, however, they share the same pains, goals, and needs as the client. Being a partner means the analytic organization shares a responsibility in socializing analytics to the organization, educating the stakeholders, and building confidence in advanced analytics so they can benefit from the results.
The most successful analytic organizations are not those with the best PowerPoint presentations, tools, or even analytics. Instead, the most successful are those that can tell the business story associated with the data and analyses, and can bridge the black box of methodologies to real-life business issues, implications, and value. These organizations are business partners to their customers, helping them achieve business objectives.
At the end of the day, analytics that helps customers meet their goals drives value. Since customers too often fear the black box, analysts are responsible for creating great analyses while also opening the black box and educating all stakeholders.