Linear trend projection
Forecast future values by fitting a straight line to your historical data.
The linear trend method fits a straight line through your historical data points using ordinary least squares (OLS) regression, then extends that line into the future.
How it works
OLS regression finds the line y = mx + b that minimises the sum of squared differences between the line and your actual data points. The slope m captures how much the value changes per unit of time, and the intercept b anchors it to your data's baseline.
Once the line is fit, TotalKPI projects it forward over the selected duration.
When to use it
- Your metric has grown (or declined) at a roughly constant rate over time.
- You expect the same pace to continue into the near future.
- Examples: monthly revenue with steady growth, a slowly declining churn rate.
Limitations
- Assumes growth is strictly linear - it will over- or under-estimate when growth is accelerating, decelerating, or cyclical.
- Sensitive to outliers: a single anomalous data point can tilt the line noticeably.
- Not suited for data with a clear curve - consider Logarithmic or Exponential smoothing instead.