Conjectures and Refutations: The Growth of Scientific Knowledge
Popper's central claim: what makes a theory scientific is not that it can be confirmed but that it can be refuted, and the growth of knowledge happens through bold conjectures followed by serious attempts to disprove them.
The book collects essays that build this argument across science, politics and philosophy, with characteristic clarity and combativeness.
For product direction the falsificationist habit is directly useful — most product hypotheses are stated in a form that cannot fail, which is precisely why they produce no learning.
Read alongside Kuhn for the sociological counterpoint and Merton for the institutional conditions that make refutation possible.
Popper writes with unusual force; the essays on demarcation and on the sources of knowledge are the essential ones.
Central argument
Popper argues that the defining characteristic of scientific knowledge is falsifiability: a theory earns scientific status not by accumulating confirming evidence but by making bold, specific predictions that could in principle be proven wrong. Knowledge grows through a cycle of conjecture and refutation — proposing daring hypotheses and then subjecting them to the most serious attempts at disproof possible. Across essays spanning natural science, political philosophy, and epistemology, Popper defends this demarcation criterion against both inductivism and dogmatic rationalism, insisting that the willingness to be wrong is what separates genuine inquiry from pseudo-knowledge.
Critique
Popper's account is largely normative — it describes how science ought to proceed — but underestimates the sociological and institutional conditions required for refutation to actually happen. As Kuhn's contemporaneous work showed, scientific communities routinely protect core theories from falsification through auxiliary hypotheses and paradigm loyalty, meaning Popper's falsificationist ideal functions poorly as a descriptive theory of how knowledge actually advances. There is also a tension in applying falsifiability as a sharp demarcation criterion: many productive research programmes, and most real product hypotheses, are clusters of assumptions where it is genuinely unclear which element a failed test has refuted.
Why it matters for product
The curator's observation that most product hypotheses are framed unfalsifiably is the direct operational problem: if a success metric can be reinterpreted after the fact or a launch can always be explained away, the team is running rituals rather than experiments and accumulating no durable learning. Applying Popper forces product leaders to specify in advance what result would cause them to abandon a bet — a discipline that sharpens roadmap prioritisation and prevents the sunk-cost reasoning that keeps dead initiatives alive. It also has consequences for team culture: a falsificationist environment requires psychological safety to report disconfirming results honestly, which is an organisational design challenge as much as a methodological one.