Belief Updating

Core Idea

Initial thinking is often wrong. Belief updating is the process of revising your understanding when new information appears — instead of defending the original idea. In product work, this matters because assumptions are incomplete, information is uneven, and reality is often different from expectations.


State of Thinking

Updating beliefs is not about changing your mind frequently — it is about changing it when necessary. The goal is not to be right from the start, but to become less wrong over time. Updating early prevents compounding mistakes.


Thinking Trace

How I approach belief updating

1. Make assumptions explicit

Before updating anything, assumptions need to be visible. If assumptions are implicit, they cannot be challenged. Examples: “there is a gap in the market”, “users prefer speed over trust.”

2. Pay attention to disconfirming signals

Look specifically for information that challenges the current belief — evidence that contradicts the assumption, observations that do not fit the current model, unexpected constraints or behaviours. This is more useful than collecting confirming evidence.

3. Identify what changed

When updating, make the change explicit: what was the previous belief, what is the updated belief. This creates a clear before → after.

4. Understand why it changed

Updating without understanding the reason leads to shallow thinking. Identify what new information triggered the update and whether the original assumption was flawed or incomplete.

5. Adjust direction accordingly

Updating beliefs should change decisions — narrowing scope, changing product direction, abandoning an idea. If nothing changes after updating, the update is not meaningful.


What changed?

v1

— skip —

Why it changed?

v1

— skip —


Evidence / Signals

Example: Helper Match SG

Initial belief: “There is a clear gap in the market for maid transfers.”

Updated belief: “This is a competitive space where differentiation matters.”

What triggered the update: discovery of existing competitors; understanding of regulatory constraints; deeper analysis of trust and process complexity.

What changed as a result: shifted from “build a new marketplace” to “find a differentiated position within constraints.”


Open Questions

  • How do you distinguish genuine belief updating from reactivity to noise?
  • At what point does frequent updating signal weak initial thinking vs healthy responsiveness?

Connections


Changelog

  • v1 — Initial node (migrated from Quartz pm-thinking, 2026-04-25)