On the difficulty of estimating opportunity cost

I recently mentioned in my best of 2024 this quote from Ryan Holiday: How to Win the War with Yourself [The Knowledge Project Ep. #208]:

The problem is that the financial upside is always clear. Opportunity costs are sometimes clear, but often not clear.

today I read this The Register article about Pat Gelsinger "resignation" from Intel. It mentions some of the opportunity costs that Intel got wrong (in hindsight, and many years later):

The Itanium was going to be the high-end architecture, so there was no need for 64 bit extensions to x86, no matter how hard Microsoft asked. AMD disagreed, forcing Intel to grumpily adopt its standard.

Atom was going to be the portable processor of choice, so let's flog off the Arm license.

I find it always extremely fascinating to read about these stories, and hopefully learn something from them. And I think, in this case, it's really as Ryan Holiday put it: the financial upside is always clear (and sometimes too optimistic) and we don't have a good approach for exploring the what-if-we-did-not-this-instead branches of the decision tree.

We'll probably never do, and then our only strategy is to be able to self-correct as soon as we detect that the upsides are not materializing as planned, or the other branches of the decision tree become more apparent or are pruned as time passes and new data or events surface.