Enterprise-wise Budgeting and Forecasting System

A leading American operator of amusement parks had a need for controlling cash and inventory for their many disparate profit centers.  At the time, they had 12 parks, with each park having multiple departments, such as rides, restaurants, gift shops , and even down to the individual hot dog stands.  While each profit center had it own P&L, they were grouped and had common inventory and purchasing.  For instance, each park might have a few dozen hot dog stands, but they ‘bought’ inventory from a common supplier, and had their aggregate profits ‘rolled up’ at the end of the period.
The company had been using spreadsheets that the local centers willed out and then forwarded to corporate for consolidation.  The problem with that was it entailed a lot of manual effort, and invariably ran into issues when new line items or new centers were added – events which occurred multiple times a week.

Out initial analysis of the situation concluded that the primary issues was one of control: control of the line items, and control of the centers meeting their reporting requirements.
So we built a SQL database which contained a list of profit centers, and their ‘roll up’ reporting relationships.  Another contained the reporting requirements and completions for each center.  Another table contained the line items unique to each center. For each period, a new Excel template was prepared with the appropriate line items.  And there was a control that ensured that each reporting center would use the appropriate template.  When a center missed a reporting requirement, it was flagged.  Inventory and sales were reconciled on the local level, and the spreadsheets were uploaded to corporate.
After uploading, the spreadsheets’ contents were melded into the SQL database – each and every line item.  After that, corporate could download synthetic  spreadsheets (created on the fly) for any period of time, and for any level of department, profit center, or park on demand.
Concomitant with the accounting processes, we made a multivariate model, using historical data for weather – temperature, humidity, rainfall, etc. for each park location.  Then we used that weather data and correlated it to attendance and sales at each of the parks.  (The weather was local data.)  From that point, it was a short jump to being able to forecast attendance, sales, and inventory levels and requirements at any profit center at any park!
Our budgeting and forecasting system gave the company an excellent management tool to manage their cash flow and inventory.   This management tool was a key event in allowing them to sell their company at a handsome profit.