‘Strong Opinions, Weakly Held’ Doesn't Work That Well

Blog Comments
1

There’s a famous thinking framework by ‘futurist’, ‘forecaster’, and scenario consultant Paul Saffo called ‘strong opinions, weakly held’. The phrase itself became popular in tech circles in the 2010s — I remember reading about it on Hacker News or a16z.com or one of those thinky tech blogs around the period. It’s still rather popular today.

This is a companion discussion topic for the original entry at https://commoncog.com/strong-opinions-weakly-held-is-bad/
2

This is a really old essay that I wrote that I now have to revisit as a result of the Data-Frame Theory of Sensemaking.

I just added a note:

Update, from April 2026: I think I’ve figured out how to do ‘Strong Opinions Weakly Held’, so please ignore this entire essay. Basically — do what the Data-Frame Theory of Sensemaking says experts do: commit to an opinion but have explicit expectancies that may be violated. This sounds trivial but isn’t, not really — for instance, how do you build the right expectancies for every situation? Go read the essay for the full model.

3 Likes
4

Ooh, this is a great connection / revisit, and you pointing it out made me realize that one strategy I’ve been evangelizing with decision documents for a while now is also Data-Frame-adjacent: explicit “revisit” criteria.

That is, when documenting some major choice, I’ve been including / asking folks to include a section about what would trigger a revision of the decision. For example:

  • we chose this architecture for X traffic; if we regularly see X+Y traffic we should revisit.
  • we’re assuming that P metric is a reasonable proxy for the hard-to-measure outcome we want; if we find them coming apart, we’ll have to switch to an alternative.

These aren’t meant to be comprehensive, but they put a stake in the ground (which can help dissenters feel heard), they remind people that decisions aren’t (usually) meant to be permanent and unchangeable, and they help educate folks’ attention about what factors matter in specific domains and contexts.

11 Likes
5

I’ve seen this sort of thinking also allows what we actually care for to be more legible. Decisions are usually in service of something larger, and going through this exact mode of thinking exercises a muscle away from decision-myopia.

So, “these aren’t meant to be comprehensive” is spot on. Merely activating that mode of thinking in a team opens a new dimension of potential movement in the future.

4 Likes