Four Steps to Successful Data Analytics
Today, businesses of all sizes use analytics. Take my favorite fruit vendor, for example. If you ask why he stopped selling in our neighborhood market, he’ll tell you it’s because shoppers at that particular market are prone to negotiation, causing him to lose money. At the market across town, on the other hand, he’s found great customers for whom he provides excellent service.
This is the heart of analytics. This fruit vendor tested servicing in my neighborhood and realized he was losing money. How many businesses today know who their most profitable customers are? Do they know who their most expensive customers are? Even if they understand this data, have they built targeted efforts to acquire more of the most profitable customers?
Credit unions, in particular, stand to gain tremendously from capturing data and using it for the segmentation of their membership.
Analytics is not pure science; it is part art as well. Credit unions that master the fine art of using analytical tools realize increased revenues and enjoy cost savings. But because scientific principles are often simpler to explain than artistic ones, let’s look at the four key steps of the scientific approach to data analytics.
- Define the business problem
Analytics begins by identifying the right problem. It requires understanding the facts, to which you have ready access, and then drawing conclusions from them to identify the business problem that needs to be solved. For example, a credit union is suffering from declining profits. By looking at the balance sheet, we realize that revenues have declined while the costs have remained constant. Through these two facts, we can identify a simple business problem – the credit union must reduce costs, or increase revenue, if it wants to have the same profitability as before.
- Propose a hypothesis
Often called an "educated guess," a hypothesis provides a suggested solution based on evidence. Researchers may test and reject several hypotheses before solving the problem. Taking the credit union example above, there may be two sets of hypothesis:
First, increase revenue by focusing on improved marketing or price reductions to increase competitive stance. Second, reduce costs by cutting the operations budget or lowering marketing expenses.
Interestingly, both hypotheses may lead to increased profitability by either increasing or decreasing the marketing budget. Of course, there are several implications of each action beyond the primary implication, and all need to be evaluated. The key element of the hypothesis-building phase is that you should have a mutually exclusive and collectively exhaustive set of hypothesis. This requires considering all the possible sets of relevant hypotheses for the situation and ensuring they do not overlap, and that together they are complete.
- Test the hypothesis.
Let’s continue with the example above and set up a test for the credit union to learn whether increasing the marketing budget would affect revenue. In this case, we would set up a test, running the existing marketing programs and calling it “Group A.” In “Group B,” we would run the increased marketing program. At the end of the observation timeframe (assume two to three months), we would measure revenue for each group to understand the differences.
Let’s assume that Group B performed better than Group A. Let’s also assume that at the same time we increased marketing, our competitors decreased marketing. Now the question becomes, was the incremental benefit driven by our increased marketing, or was the benefit due to the fact that competitors reduced their marketing? Assimilating all possible and relevant information is extremely important in order to reach a good decision.
Scientists have been putting the above techniques to work for a long time. Businesses, on the other hand, are just beginning to see the value of scientific data analytics.
When you create an environment in which you are constantly testing, you are constantly learning and evolving. With a commitment to the scientific process and a systematic approach to testing and learning, credit unions can evolve, as well, increasing knowledge and creating truly valuable and sustainable products.