What to do to avoid the deemed dividend trap.
Research tax-efficient structures to facilitate real estate investing.
Using the Income Tax Act to avoid tax.
Utilizing the butterlfy.
Navigating through the delicate nature of non-arms length transactions.
Applying Statistical Analysis To Forecast Strategy
The exercise of analyzing data from operations gives productive and valuable insights and drives future strategic decisions. A common approach to statistical analysis of data to determine strategy lies in revealing correlations between revenue levels and their respective drivers.
Our case study involves assessing the results of a strategic initiative deployed by First Mattress and Fittings, a hypothetical bed mattress and frame manufacturing entity suffering from years of lackluster sales and declining market share.
The Board of Directors has decided to insert a new management team, and as it’s new strategic direction the team decides to employ a strategy of cost leadership. Previous surveys to it’s target market nationally have revealed that its’ products are of sufficient quality, and, because time is of the essence, the team decides to forfeit additional market studies to determine revenue drivers and employ the cost reduction strategy.
Information tracked over the 5-year period were:
- Monthly tabulation of sales made
- Monthly national market share calculator
Collaborative efforts with the board resulted in a consensus on intended results over a 5-year strategic campaign of:
- Yearly market share increase
- Net income levels of 20% EBITDA
- Does the data provide sufficient metrics to employ accurate forecasting efforts for future campaigns?
- Is there room for future growth under the current strategy?
- What is the optimal time frame for strategic campaigns?
That was 5 years ago, and the team has assembled its’ data optimization department to determine insights that can be used to drive future strategic decision.
Comparing the monthly sales to time, the charting reveals more sales in the summer months than the winter months. A strong advertising campaign employed over the 5-year period shows strong improvements in market share at the outset of the campaign; however, the increase in market share declined at an increasing rate as the campaign progressed, suggesting a finite life to product interest and consequent need for improved offerings based on market research.
The sales chart with time as the x-axis and sales levels as the y-axis revealed an upward-sloping line from left-to-right with acuteness dissipating as time wore on. Testing the correlation r showed a value of 77.9% (after discarding outlier data points) confirming that the strategy of cost reduction resonated with consumers and resulted in increased unit sales of mattresses and frames.
How can this information be used to drive accurate forecasts and strategic agendas? The quantitative information needs to be reconciled with qualitative, executive considerations (“gut feelings” in the vernacular), which enhance the quality and probability of success of the strategic venture.
The following drivers should be selected to calculate co-relation with sales, which should then used to build a multi-variate forecasting model.
- Population density
- Weather at location (average temperature)
- Average age
- Income level
Subsequent trend and seasonality insights from the 5-year campaign are integral to optimizing the quantitative and qualitative contributions to the forecast, and should be studied appropriately to determine any additional insights for the forecast.
The correlations between each independent variable and the calculated sales level will enable us to build a model to apply in the forecasting exercise. This model is then used to predict, first, sales by region, and then component assets required to facilitate sales levels, such as manufacturing equipment, staffing requirements, operational space needs, and administrative support.
The benefits of data to strategic forecasts have merits, but only to the extent that those using the data can apply the following:
- Is the data relevant to what we want to forecast?
- Can it be used to accurately apply to our strategic forecast
- Can it provide us with accurate information on the drivers influencing what we want to forecast?
Understandably, forecasting is not a science, but the utilization of complete, accurate, and practical data can bring us closer to this optimal stat
Nicholas Kilpatrick is a partner at the accounting firm of Burgess Kilpatrick. He leads the firm’s consulting and strategy practice and works with companies to enhance their Analytics, Forecasting , and Data Optimization functions. The practice’s focus includes quantitative forecasting, corporate and unit strategy and planning. Please visit our website at www.burgesskilpatrick.com or on Facebook at www.facebok.com/BurgessKilpatrick for more information on our firm.