Start Leveraging Analytics Today!

Financial forecasting is a combination of art and science. There is no arguing that.

 

However, financial planning has typically been weighted toward art, with finance teams heavily leveraging their best judgments to forecast, often without much scientific or analytic support. Certainly, a business leader’s experience is a valuable contribution to an accurate forecast and building alignment with its vision. But ignoring the more scientific aspects of the equation can often leave us blindsided.

 

Many organizations already house a wealth of data. Applying simple statistical techniques to analyze this data will help isolate insights that leaders can often miss.

 

Using the example of a utility company’s electricity rates over 10 years, you’ll see how different analytic techniques can reveal hidden insights.

Start Leveraging Analytics Today

Looking at the above chart, most of us will only see a steady growth of electricity price rate over the years. The graph’s nooks and crannies simply look like random noise. However, those crannies hold a wealth of information and insights that we would have missed if we stopped analysis here.

 

We can discover many things by introducing a simple statistical model, often taught in first year college stats, called time series forecasting (TSF). This statistical model observes trends across time and explains the “why,” using three components – trend, seasonality, and outliers.

 

The following graphs display the insights that this model found without any manual inputs. The model identified a smooth, steadily increasing growth trend that better quantifies the year-over-year growth. The model also found a clear pattern of seasonality, which could have been easily mistaken for pricing growth if undiscovered.

 Start Leveraging Analytics Today

Start Leveraging Analytics Today

Additionally, the technique identified outliers that were unexplained by an overall trend or seasonality analysis. These may be one-time events such as a city-wide blackout or specific weather incidents. Although they are irregular, analyzing these outliers and uncovering their underlying causes can help businesses prepare for unaccounted risks and discover opportunities to improve customer experience.

 Start Leveraging Analytics Today

Electricity rate forecasting is critical to a utility company and ignoring the analytical analysis would have failed to capture extremely important insights for the business. Without these analytic tools, organizations would have missed opportunities to reengineer business operations to better prepare against irregular incidents or improve pricing models to better-fit consumer pricing sensitivity.


Time series forecasting is just one example out of a large array of analytic techniques, many of which are easy to understand and simple to implement with readily available software. For instance, Microsoft Excel has statistical functions such as TSF built into its software. Even neural network analytics, often positioned as cutting-edge, can be built into Excel spreadsheets.


These models can be applied to any industry and can be implemented within systems such as corporate performance management solutions that act as central data warehouses. At CCH Tagetik, we have a brand new analytical platform that has already started utilizing statistical modeling for industries such as healthcare, where time series forecasting and regression models are used to predict medical measures such as ventilator volume, intensive care units, and surgery counts to improve forecasting accuracy on revenue and expenses.

 

We often see innovative techniques such as artificial intelligence and machine learning in the headlines, but such technologies all began as basic statistical models as well. Solutions employing algorithms based on machine learning can become black boxes if an organization lacks the analytical experience to infer the output correctly or govern the data input appropriately.

 

Artificial intelligence will definitely become more prevalent in finance in coming years, but there is no reason not to start incorporating more advanced analytics into your planning processes today. Begin by incorporating simple statistical modeling and move towards more sophisticated tools based on machine learning as you gain more competency. By doing so, you will see rewards right away while ensuring long-term success in adopting a more analytical forecasting process.

cch tagetik solutions

Share this post!