You probably have to go quite a long way into answering "is it feasible" type questions, before you can answer "is it cost effective" questions. Maturing a technology will typically take many steps, each with exponentially increased costs. Finding the money for the initial (cheapest) steps may be a lot easier, especially since the pay-off may prove to be higher than expected.

Research is akin to gambling. You cannot predict which bets will pay off, but if you can win on average, it's worth betting as many times as you can afford.