Moving average projection
Smooth out short-term noise and project the recent trend forward.
The moving average method projects future values by computing the average of the most recent data points and carrying that value forward as a flat forecast.
How it works
TotalKPI calculates a simple moving average (SMA) over a trailing window of your most recent data points. The average of that window becomes the projected value, held constant across the forecast period.
The window size adapts to your dataset: larger datasets use a wider window to capture a more stable trend; smaller datasets use a shorter window to stay responsive to recent changes.
When to use it
- Your metric is noisy or volatile and you want a smoothed-out view of where things are heading.
- You don't expect a strong upward or downward trend - the metric is relatively flat or range-bound.
- Examples: daily active users with day-of-week variation, support ticket volume.
Limitations
- Produces a flat forecast - it does not capture an upward or downward trend. If your data is trending, use Linear trend or Exponential smoothing instead.
- The window size is fixed automatically - very short datasets may produce less reliable averages.
- Reacts slowly to sudden shifts in the metric because older values remain in the window.