Analytics runs all business, yet it’s so often obfuscated. To make it real is a journey of discovery – first, establish a hypothesis, second, build an understandable context and third, act after you have continued to question the problem to conclusion.
It’s a Tsunami! Data is estimated to be doubling in size every 2 years; data analytics has never been so critical in history. Today’s CFO is more a customer focused strategic role and data driven leadership than compliance-driven and optimization.
How is the market reacting? Part of FP&A’s effort to Keeping Analytics Real is having real and true data. In the landmark FP&A Empowerment Survey 2017, over 200 FP&A and Finance professionals agree that only 21% of their time is spent on High-Value Activities compared to the 48% they aspire to spend. Utmost in that includes business partnering; strategic support; customer facing activities and driving actions. (The FP&A Empowerment Survey 2017 is closing soon – make sure you have your say!) Time is the biggest challenge for all Finance professionals. The solution is to make data analytics more effective in 3 easy steps:
Step 1: Hypothesis
What do you expect?
Before starting the analytics, always ask the question: “What do we expect the results to be?” At first glance, this may seem counter-intuitive; yet there are various reasons why you would want to ask the question. Part of keeping it real is to first establish a hypothesis i.e. What does our customer really want? Too often, I’ve seen CFO’s just ask for a report; when they get the report there are several additional questions. Time is everything and that old process wastes valuable time. KEY: What you are establishing is a base line or a hypothesis. It helps to get to the output quickly, but without the hypothesis, you are just reporting!
A typical example is when you’ve been asked to supply the top ten customer performance over the last year – a simple task.
But what they really want to know is how to have the top customers in each country been performing, as they believe the competitors are targeting your top competitors. The hypothesis is now: Are my top customer by the country being targeted? – which may result in very different analysis then just looking at the top customer list. You must test assumptions as for whether it’s true or false.
The other great thing about asking that question is that it gives your insight into your customer’s bias. Everyone has a bias, no matter how independent you think you may be. Results can often be interpreted in different ways depending on your preconceived ideas. Once you have established a hypothesis, the focus switches to either passing or failing that hypothesis, saving time, resources and angst!
Step 2: Build Context
How should it be seen?
It easy to misinterpret data in isolation. As an example, if you see a table with single data points, it’s impossible to understand its relevance. You must have some context. Sometimes, it can be as easy as adding a Sparkline to the analysis. This helps the user visualize the prior history. A crutch that must be avoided is “Pie chart mania!” the overuse of pie charts to build context. Pie charts maybe great visuals to articulate proportions, but humans struggle to interpret the rate of growth/decline by the size of the circle! (i.e. if you have 2 pie charts next to each other, it’s difficult to tell if they are growing or declining in size). Even when it’s obvious, it difficult to tell the rate of growth/decline. Build context with line charts; they are far more effective at depicting growth or decline.
Continuing the analogy, if the hypothesis is: My top customer by the country being targeted, then a simple analysis would be to compare the growth rates from the top customer to the average growth rates of say the next 10.
Step 3: Action
What’s the problem and what should be done?
Seems simple but it’s amazing how many times finance teams do analysis without know what problem they are solving! Or they fall into the trap of looking at it from a Finance perspective vs. a business perspective. It’s critically important to understand both the problem and to the context of the person asking for the analysis. You need to place yourself in the shoes of your customer – understanding not only the business environment but how he/she thinks about the environment. Once we understand the problem in more detail, then the corrective actions are more obvious. In the case of our customer being targeted – the real problem we are trying to solve is How do we defend our customers from being targeted by competitors? That may require more detailed driver analysis. (e.g. How have sales opportunity been performing over the last few months? What are the reasons for the decline: lost on pricing; a configuration or is the pipeline declining?) Once we understand the underlying problems, the action should be evident, but we need to understand the problem before we can act.
Conclusion
Try applying these 3 relatively simple steps the next time you have occasion and watch the process move with alacrity and the comprehension appear to your customer. Go ahead, take a risk to turn that customer frown into a smile! Keep it real.