In a process of asymmetry, we need to acknowledge the role of the one-way door. Imagine you're in a room with five doors. Once you pick one, and go through, you can't come back and revisit your choice. Let's call one of those doors the 'First To Market' door.
The way that linear Development decision-making behaves in many places now is analogous to that one-way concept. The phase I, phase II, etc., are chosen, and success or failure ensues.
However, this is largely based on the premise that, if the drug doesn't do something we want it to, or it surprises us in some other way, we'll have enough information to pivot. However, given the narrow focus deployed in early phase, we will not. The clues that we will want, with which to make a different decision, will not be there. The clean and easy path will yield little texture, no dust kicked up that might be interesting. Absence of evidence will, to the decision process, look exactly the same as evidence of absence.
If, in that room, you were standing with your competition, your speed through that door might have felt like an advantage. Your 'decision' would have been based on any knowledge you had at the time, and your prediction of confidence and opportunity.
Let's consider that one of the other doors was the 'Best To Market' door. Is one automatically better than the other? You know that the consequences of the decision to go through this door are different than the First To Market - perhaps a different value proposition is explored, a different dose range, different patient population.
The thing of it is: your decision of First or Best has to be made before you go through the doors. The Best door, of course, might be able to pivot back to First, but it probably won't work the other way. Decisive hesitation would introduce optionality that your hesitant decision to go fast would not.
The other three doors may even be unmarked, but you know they are all different - they're also one way, but you know that they will teach you different things. The question for you, standing next to your competitors, is how much time to spend in the room before you head for a door...
In a linear Development path, the capital D decision is set early. The absence of evidence for any alternative path, which is a consequence of the fast and narrow path, is your Decision. Your TPP may be varied by a degree or two, but generating a different one will be lost to your process.
In an exploratory and confirmatory path, you will decide to Decide later. Setting yourself up for a future decision forces a change in behaviour, towards opportunity seeking and managing uncertainty - addressing unknowns and knowns. Your first decision (what to explore) is not the same as your second (what to prove). Your second Decision is significantly better informed than your competitor.
Key to Asymmetric Learning is that all decisions are not the same. Knowing that key decisions lie ahead, rather than being needed today, right now, helps change the approach. Deciding to enhance a future decision is key. The alternative is to rely on knowledge that already exists in your room, and we know the statistics on that approach.