The request to model extreme share cannibalism in a DCF is one of those quiet moments where the math stops being theoretical. When one product doesn't just compete with another but actively consumes its predecessor's revenue, the typical growth assumptions fall apart. You can't simply layer a growth rate on top of historicals and call it a day. The person asking this question is already ahead of most analysts because they've spotted the real issue: the model isn't about the new product's upside, it's about the speed and magnitude of the old product's decline.
Practical terms matter here. If you're building a DCF for a company where Product B is eating Product A's share, the first thing to model is the crossover point, not the terminal value. When does the cannibalized product stop generating meaningful cash flow? What does the combined margin profile look like during the transition? The risk isn't that the new product fails; it's that the old product's decay outpaces the new product's ramp. That's where most models get optimistic. They assume the new product grows into the gap, but they underweight the fixed costs that don't disappear when the old product shrinks. The answer isn't to ignore the cannibalization, it's to build a separate revenue bridge that maps the decline of the old and the ascent of the new on a quarterly basis.
The practical takeaway for anyone doing this work is to stop treating the DCF as a single forecast and start treating it as a scenario tool. Run the bear case where the new product captures share slower than expected, and the old product lingers longer because customers are sticky. Run the bull case where the new product accelerates because pricing power holds. The point isn't to pick a number you like, it's to see which assumptions break the model. If the valuation swings wildly based on a six-month delay in cannibalization, then the margin of safety needs to be wider, not narrower. That's the discipline the original poster is circling around, and it's worth naming it directly.
So, the concrete move is this: model the cannibalization rate as an explicit assumption, not a hidden one. Put it on its own tab, with a chart that shows the revenue crossover and the cash flow crossover separately. They won't happen at the same time, and that gap is where the risk lives. If the new product's gross margin is higher but the old product's fixed costs are still dragging, the free cash flow will dip before it recovers. That dip is the real test of whether the investment thesis holds. If you can't stomach that dip in the model, you shouldn't own the stock. That's the answer to the question, and it's a lot more useful than a discount rate tweak.