Trend based
DCFForecasting by extrapolating historic growth rates or margins forward, with no explicit operational build.
A trend based forecast takes what the business has been doing and assumes it continues, perhaps fading toward a lower rate over time. It is the quickest method and the default when disclosure is thin.
It is legitimate for a stable, mature business in a stable market, and for the later years of any forecast where pretending to model operational detail a decade out is false precision.
It is dangerous anywhere something has changed. Extrapolating the growth of a company that just lost its largest customer, or the margin of one that benefited from a one off input cost windfall, produces a confident forecast of something that will not happen.
The honest use is as a cross check on a driver based build. If your detailed forecast implies growth far above the historic trend, that is not necessarily wrong, but it is a claim requiring a specific reason.
Worked example
A business grew 7%, 6% and 8% over three years, so 7% is extrapolated forward.
That is reasonable until you learn its largest customer, 18% of revenue, did not renew. The trend is now describing a company that no longer exists.
Trend extrapolation is legitimate for the back years of a forecast, where operational detail a decade out is false precision, and dangerous anywhere something has changed.