I used a 12% discount rate, which is, from what I've read, in...

I used a 12% discount rate, which is, from what I've read, in the middle of the range commonly used by pharmaceutical companies in drug npv calculations.

Much higher discount rates can be used in a dcf in other scenarios, for example if you are valuing a tech startup where the uncertainty in your assumptions and forecasts is often sky high.

Tech is actually my area of expertise and where I have used dcf modelling before. I have to say that pharmaceutical modelling is a dream compared to this - the simply enormous amount of data available and the significant number of factors you can estimate (relatively) accurately is unlike any other business area I'd say, where the uncertainties are normally far greater.

There is certainly more upside in the model I did, not in the discount factor so much, but in other areas like market penetration, how many additional indications the drug might target, and how long patent protection could be extended.