Self-driving cars, Augmented Reality, Advanced Analytics and Artificial Intelligence (AI), all have two things in common; the use of advanced technologies to 1) eliminate or streamline many of today’s manual, particularly mundane tasks and 2) produce better results. The latter may (ultimately will) be fewer car accidents, more accurate forecasts and if you are a sports fan, a championship team.
While Insurance and Banking have used Advanced Analytics and more recently AI effectively to detect fraud, assess risk or predict equity prices, every industry is attempting to use advanced analytics to gain a competitive edge, every domain now led by Digital Marketing is using advanced analytics and AI to further enhance that advantage. Facebook tracks every post you read and every post you make, using Advanced Analytics and AI to determine which post or article to deliver to you next to retain your engagement allowing them to sell more ads. Really there is no ‘fake news’, the news Facebook chooses to deliver to you – is all based on what has been determined, using Advanced Analytics and AI to retain your engagement, not about its authenticity to sell more ads.
The Office of Finance will not be left behind, typically laggards in adopting technologies the Office of Finance will be (is) creating forecasts automatically using Advance Analytics creating more accurate forecasts with minimal manual efforts. As the applicability of AI to the Office of Finance evolves, decisions will be made by AI effectively sending the Financial Analyst the way of the Dinosaur.
The Future of Forecasting is here today; creating dynamic forecasts using Predictive and Advanced Analytics without the need for teams of Department Managers and Finance Business Partners to manually develop forecasts in a costly time consuming process.
Variance Commentary? No problem, Natural Language Processing tools are here today and can seamlessly create commentary. Perfect? No, but it is much easier and faster when the Draft commentary is already written.
The number one reason organizations don’t create rolling forecasts and more frequent forecasts is the level of effort.
The concept of Driver Based forecasting has existing for decades, during the late 1980’s we were converting hospitals to driver based forecasts using projected inpatient and outpatient volumes to ‘drive’ departmental specific volumes and ultimately ‘drive’ staffing requirements and medical expenses. The result was a streamlined annual budgeting process (monthly reforecasts weren’t even a thought), more accurate forecasts and most importantly if the forecast was wrong we knew it was either volume or acuity driven (sicker patients).
Prior, forecasts were mostly wrong with enormous amounts of time spent analyzing the variances to reach the same conclusion 99% of the time; the Department Manager didn’t have the right information to forecast correctly – they didn’t know the forecasted volumes or they were wrong. By closing the loop between forecast patient volumes and the departments providing the care we saved significant amounts of time creating the forecasts, analyzing and explaining the variances and simply not aggravating good people.
Really it was easy, after Finance Developed high level Patient Admission and Outpatient visits we met with our Physicians to validate our assumptions; mostly to understand three things 1) what medical advances should we anticipate that would impact volume; for example an inpatient procedure that would move to outpatient 2) physician referral patterns and 3) what concerns if any about our competition
Once high level volumes were established, a waterfall process based on historical trends ‘drove’ the volumes for every patient care and non-patient care department from the ICU to Admissions, Billing and Food Services. Staffing metrics were developed for how many hours we needed to process each patient claim to the number of hours to feed an inpatient which drove headcount and staffing expenses. The Department Managers job? Reconcile and justify the number of staffing hours needed to the number of staffing hours required to provide minimum coverage – which in turn fostered and drove conversations among departments on cross coverage, cross training and ultimately reduced staffing costs
The teams were focused on execution, not entering numbers in a spreadsheet.
For more information on the Advanced Analytics and Natural Language Processing for the Office of Finance;
Advanced Analytics, Simulation and AI;
Natural Language Processing: