Scattergraph method
Scattergraph method estimates a cost function by plotting historical cost against activity and drawing a line of best fit by eye; it is less precise than regression but spots outliers before statistical estimation is applied.
FrameworkCost estimation
See it move
Historical cost and activity pairs are plotted, then a straight line is drawn through the cloud by eye rather than computed. Where the line crosses the vertical axis estimates the fixed cost; its slope estimates the variable cost per unit. Using every data point rather than two extremes makes it less distorted than the high-low method, and the plot itself reveals outliers before any regression is applied.
The formula
Variables
- Total cost estimated from the visually fitted line at a given activity level (€)
- Fixed cost component (y-intercept of the line drawn by eye through the scatter plot) (€)
- Variable cost per unit of activity (slope of the visually fitted line, read as ΔCost ÷ ΔActivity) (€ per unit)
- Activity level (cost driver quantity) (units)
The line is positioned by eye rather than computed statistically; the primary value is visual detection of outliers before applying regression analysis.
Check yourself
A transport manager plots 24 months of fuel cost data against kilometres driven on a graph and draws a line of best fit by eye. Compared with applying the high-low method to the same 24-month dataset, what is the principal advantage of the scattergraph approach?