4.9 - Sales Forecasting
The meaning and purpose of sales forecasting
Sales forecasting involves predicting the amount of sales revenue a business expects to generate over a specific timeframe using quantitative methods.
Forecasts rely on data such as recent sales patterns, market research, the state of the industry, and broader economic factors. These predictions are often shown as time series data, which highlight trends from historical sales information.
How sales forecasting supports business functions
- Human resources planning - Indicates when additional staff might be required to handle expected increases in demand.
- Financial forecasting - Assists in creating cash flow and profit projections by estimating future revenue.
- Production and inventory management - Helps schedule manufacturing and stock levels to match anticipated sales.
- Strategic decision-making - Reveals market trends, allowing businesses to spot opportunities like new markets or threats such as emerging competition ahead of time.
How to calculate moving averages and use extrapolation
Moving averages and extrapolation are key quantitative tools for identifying trends in sales data and making predictions.
Calculating moving averages
A moving average smooths out fluctuations in data to reveal the underlying trend. These fluctuations often stem from seasonal, cyclical, or random factors. It is commonly used for quarterly data by averaging groups of four data points.
The process involves two main steps:
- Calculate a moving total, such as the sum of sales for four consecutive periods.
- Calculate the centred average by dividing the moving total by the number of periods (e.g., 4 for quarterly data).
Where:
- Sum of data points in the group = Total for the selected consecutive periods
- Number of data points = The size of each group (e.g., 4 for quarterly)
Using extrapolation for predictions
Extrapolation extends identified trends from past data to forecast future sales, either in units or revenue. It assumes that historical patterns will continue, allowing businesses to plot a trend line and project it forward.
Worked example - Calculating moving averages
A business records quarterly sales (in £000s) as follows: Q1: 120, Q2: 150, Q3: 180, Q4: 140, Q1 (next year): 130, Q2: 160. Calculate the four-quarter moving averages for the available periods.
Step 1: Identify the values
- Quarterly sales: 120, 150, 180, 140, 130, 160 (£000s)
Step 2: Calculate moving totals
- First group (Q1 to Q4): 120 + 150 + 180 + 140 = 590
- Second group (Q2 to Q1 next): 150 + 180 + 140 + 130 = 600
- Third group (Q3 to Q2 next): 180 + 140 + 130 + 160 = 610
Step 3: Calculate moving averages
- First moving average: 590 ÷ 4 = 147.5 (£000s)
- Second moving average: 600 ÷ 4 = 150 (£000s)
- Third moving average: 610 ÷ 4 = 152.5 (£000s)
Step 4: Interpretation
The moving averages show an upward trend in sales from 147.5 to 152.5 (£000s), smoothing out quarterly fluctuations.
The types of variations in sales data
Sales data often includes variations that cause deviations from the main trend. Understanding these helps in creating more accurate forecasts.
Seasonal variations
Seasonal variations are predictable, repeating changes in sales revenue over a year, often linked to specific times like holidays or weather. They are calculated as the difference between actual data and the trend line value, which can be shown in absolute figures or percentages.
Cyclical variations
Cyclical variations are repeated ups and downs in sales tied to the broader business cycle, such as economic expansions or downturns. These can span more than a year, unlike seasonal ones, and are influenced by overall economic conditions.
Random variations
Random variations are sudden, unpredictable shifts in sales caused by unforeseen events.
Examples include:
- Natural disasters, extreme weather, conflicts, health outbreaks
- Company scandals or negative publicity
Benefits and limitations of sales forecasting
Sales forecasting provides valuable insights but has inherent constraints that businesses must consider.
Benefits of sales forecasting
- Trend identification - Smooths variations to highlight underlying patterns in sales data.
- Uncertainty reduction - Gives managers greater confidence in future planning by anticipating changes.
- Planning support - Forms the foundation for budgets, strategies, and resource allocation across departments.
- Opportunity and threat detection - Allows early recognition of market shifts for proactive responses.
Limitations of sales forecasting
- Short-term accuracy - Predictions are often reliable only for brief periods, with longer-term forecasts prone to error.
- Assumption of continuity - Relies on past trends persisting, which may not hold in changing environments.
- Data requirements - Needs high-quality, expensive-to-gather information that might not always be available.
- Challenges for diverse businesses - Less effective for firms with varied products or rapidly shifting customer tastes.
- Qualitative oversight - Ignores non-numerical factors like competitor actions or consumer behaviour changes.