Rolling Forecasts

With company valuations in the dumpster and credit markets tight, the economic crisis is forcing hard looks at revenue sources and spending across the board.

The economy aside, old ways of doing business are continuously challenged by competitors with innovative business models enabled by the internet. As well, the old way of thinking — that you run a company through the budget process — is on the way out in favor of new approaches such as continuous planning and rolling forecasts.

Few finance organizations are ready to dump budgets per se. However, many are looking critically at their spreadsheets or canned budgeting applications asking how they can deliver more immediate and responsive financial plans.

The answer is rolling forecasts where actuals are integrated with plan time periods and the forecast horizon extends beyond the current fiscal year. However, it's not easy to do.

Rolling forecast basics
The complexities of importing actuals
Types of variance analysis

Rolling forecast basics

Budgeting defines the basic structures of financial plans. However, with budgets outdated within a couple of months of publication, developing a monthly or quarterly rolling forecast is the only way teams can update business plans and stay tuned with the marketplace.

• The forecast cycle — A major problem in implementing rolling forecasts, especially if the cycle is monthly, is time. Forecasting often consumes the finance team and pre-empts collaboration because the window between the month end close and producing a forecast is too tight.
  
  Forecasting is faster and data volumes are reduced where the planning structure includes modeled elements: revenues and variable costs driven by operational drivers or quantifiable activities of the business — e.g. volume measures such as units, transactions, subscribers, customers, call levels, hours, installations and the like. It's called driver-based planning. Modeling using activity drivers is incorporated into Alight's basic unit/rate/amount architecture.
• Integrating actuals for rolling forecast — Variance analysis requires that all elements of the plan, budget or forecast, be "frozen" when comparing to actuals data — that is, the budget or forecast data for comparisons are exactly as originally presented to management.
  
  For rolling forecasts, however, many line items where calculations cross over time periods should, in fact, be based on recent actuals. For example, forecast sales should be driven in early months from actual prospects or orders in the sales pipeline. Quarterly commission payments should be computed from actual commissions booked plus forecast commissions. Alight handles this "look back at actuals" condition for rolling forecasts.
• Scenario analysis for rolling forecasts — While budgets may be "frozen" for the year, scenario analysis during the rolling forecast process is important — especially for testing impacts of alternate timing of milestones such as launching a product, changing pricing or moving into a new facility. Alight incorporates a broad range of tools for constructing and comparing scenarios including automating the rollover of underlying time periods.
• [ download white paper: Application Requirements for Rolling Forecasts ]

The complexities of importing actuals

Financial analysis and re-forecasting starts with comparing actual financial and operating results with plan. The trick is getting all of the parts imported from multiple database systems and lined up apples-to-apples in a rolling forecast format and for variance comparisons.

• Level of detail — Managers frequently plan at a lower or different levels from accounting structures — for example, a sales manager may plan sales by product by channel but the general ledger delivers only sales by product — the real sales detail is in another database. Alight supports importing from multiple disparate databases at any level of detail, not just GL totals.
• Underlying units and rates — Granularity of detail is critical. Managers who plan using a unit/rate/amount architecture want to see that level of granularity with actuals data. Alight supports importing any combination of actual units, rate and amount for all P&L and balance sheet items. Alight also back-calculates missing elements — for example, if you import sales units and dollars, Alight calculates average price. If you import dollars and average price, Alight calculates the units.
• Operational data — Importing actuals should incorporate operating metrics as well as general ledger amounts — especially when financial plans are activity based with assumptions about activity levels and productivity. Alight supports importing operating data at any level with modeling or back calculating of ratios and metrics. For example, if you import billed consulting hours and consulting staff headcount, you can model and calculate consulting staff utilization.
• [ download white paper: Integrating Actuals into Financial Plans ]

Types of variance analysis

Variance analysis is the rubric for comparing actual results with plan — whether budget or rolling forecasts. There are three types:

• Traditional variance analysis — It works like this: compare actual amounts at the natural class account level to budget or forecast with a column that computes the dollar or percentage variance. Alight does this for all combinations of time periods — month, year to date, full year, etc.
• Not so traditional variance analysis — It should work like this, but usually doesn't: compare actual units, rates and amounts at the line item level to budget or forecast with columns that compute variances for all three data types. Alight does this.
• Causal analysis — Where actuals and plan line items include units and rate as well as dollar amount, you may compute a causal analysis variance. This variance type calculates how much of the total dollar variance is due to higher or lower units (the volume impact) or a higher or lower price/cost (the rate impact). Alight automatically computes volume and rate impacts for all revenue, expense, headcount and balance sheet line items that incorporate units and rates.