Growth rate (CAGR) projection

The growth rate method uses your data's Compound Annual Growth Rate (CAGR) to project future values as compounding percentage growth.

How it works

CAGR is calculated from the first and last values in your historical data:

CAGR = (end_value / start_value) ^ (1 / n_periods) - 1

Where n_periods is the number of time steps between the first and last data points, expressed as a fraction of a year.

The projection then applies this rate repeatedly from the last known value:

projected_value = last_value × (1 + CAGR) ^ t

Where t is the number of periods ahead being forecast.

When to use it

• You think about your metric in terms of percentage growth per period (e.g. "growing at 15% per year").
• Your data shows exponential-style growth - each period's increase is proportional to the current value.
• Examples: revenue with consistent growth rates, compound interest, subscription counts scaling multiplicatively.

Limitations

• Highly sensitive to the first and last data points - outliers at either end distort the calculated rate significantly. Consider whether your start and end values are representative.
• Assumes the historical growth rate will continue unchanged, which is often optimistic over long horizons.
• Not suitable for metrics that are flat, declining, or negative - CAGR is undefined for zero or negative starting values and produces misleading results for declining series. Use Linear trend for declining metrics.