Successful Forecasting Techniques

Recently I was asked by CFO Talks to present a webinar on “successful forecasting techniques”.

I accepted with some trepidation. Let’s face it, in normal stable times, successful forecasting can be challenge.

In times of uncertainty, it can be quite daunting.

Which of course, is no reason to avoid it or step away.

Irrespective of business conditions, forecasting is an essential element of successful business performance.

Forward planning – in the form of forecasting –  is the only intelligent way to attempt to predict the future.

And as it is predictive, it will be subject a a margin of error.

Of course we need to define what it is that we are forecasting – in a corporate environment it could be revenues, profits, working capital, cash flows, dividend capability, sustainable growth, and a variety of returns.

My point of departure is cash flows and the integrity thereof  – which means the process must enable the validation of the forecast..

So what do we need to consider – at a macro level.

1. Presence (or absence) of history
2. Industry – business tempo – differences between retail, banking, mining, contracting
3. Industrial Action – unpredictability
4. Political activity
5. general economic conditions – inflation, GDP growth, interest rates, commodity cycles,

exchangerates, fuel prices

1. market share
2. product range
3. competition
4. human resources
5. available capital

Historically, forecasting has taken various forms:- 

a. traditional historic budgeting (various techniques) – static budgets
b. forecasting and rolling forecasts – combination of actuals and budgets

More recently – and given its significant advantages – financial modelling has become the preferred process.

Whatever methodology is adopted, the following key elements must be present:-

1. qualitative reliable research – defining the forecast drivers
2. flexibility – able to cater for rate of change in business conditions
3. adaptability – able to cater for changing business structures
4. structured and easy to operate
5. modifications must automatically flow through to all related issues

Modelling advantages                                                                                                                                 

1. Models are dynamic – flexible and easy to change
2. Effective models facilitate “what -if” analysis – forward changes based on variable inputs
3. Backward changes using Excel’s “goalseek” functionality
4. Completely integrated financial statements comprising – Income statement – Cash flow

statement – Balance sheet

5. Modelling process validates accounting accuracy of financial statements
6. Easy adjustment of time periods make models reusable

Modelling disadvantages

1. Time required to build the model given the comprehensive and holistic nature of a model
2. Validating and verifying the primary input assumptions – model drivers – qualitative and

quantitative research

Conclusion

1. all forecasts are predictive – and as such will be subject to a margin of error
2. all forecasts are only as good as the primary assumptions – the underlying research
3. forecasts must be able to address the rate of change in business – must be flexible
4. budgets are a form of forecasting – but are static – immediately out of date
5. rolling forecasts are a combination of actuals and budgets – and have integrity concerns
6. financial models are flexible, structured and permit easy “what-if analysis”
7. financial models integrate revenues, costs, assets, liabilities and the methodology proves the

validity

1. effective models are easy to use, easy to understand and easy to verify
2. financial models are the most effective form of forecasting

Colin Human CA(SA)