Hack:
No crystal ball to predict budgets

Managers in medium to large companies spend an incredible amount of time planning operations and finance. With an increasing complexity in businesses and exponential numbers of customers and products, this financial exercise can take months and tends to be very political and disconnected from operations.

Much research has been published about methodologies for better budgets, like:

• Activity-Based Budgeting,
• Balance Scorecards,
• Time-Driven Activity-Based Costing,
• or Rolling Forecasts.

This is also a milestone for every degree in Business, and exams for CPAs (Certified Public Accountant), CMAs (Certified Management Accountant), PMPs (Project Management Professionals) or MBAs.

However, there hasn't been any major breakthrough in how technology could help managers in this area. Most planning software packages try to either facilitate the process through workflows or concentrate on the data collection and processing through data warehousing.

A radically new approach would leverage computer systems to automatically calculate a baseline budget for every product, customer and distribution channel and allow managers to only spend time correcting or simulating the scenarios that would really have an impact. Through the power of software like SAP Predictive Analytics and HANA, leveraging data history and mathematical models, more time would be spent on analysis as opposed to number crunching and Excel chaos, thus enabling managers to add value to the process as opposed to suffering through it.

A typical example could apply to the cosmetic industry. With hundreds of brands, product families and products, thousands of customers in every country and multiple distribution channels (in-store, duty-free, web, etc.), the combinations to be planned can be counted in millions or billions. The usual workaround consists in using groups (of products or customers) or averages (sales prices). Instead, with predictive solutions, sales quantities could be calculated based on history and applicable product life cycle. Then, managers could spend more time simulating the impact of price changes, new product introduction or customer / product termination.