Small Wins

Definition

Small wins are minor, achievable victories that matter far more than their size suggests, because each one changes what people believe is possible and sets up conditions for the next win. Duhigg cites organizational psychologist Karl Weick's description of them as "miniature experiments that test implicit theories about resistance and opportunity and uncover both resources and barriers that were invisible before the situation was stirred up" — they don't chain together in a neat, predictable line toward a goal; they accumulate unpredictably, each one shifting the landscape for the next.

In the Book

Duhigg's clearest example is the 1970s gay-rights movement: after years of defeats trying to repeal discriminatory laws, an American Library Association task force won one narrow, unglamorous victory — getting the Library of Congress to reclassify books about gay liberation out of the "Abnormal Sexual Relations" category. That single small win spread as news, fueled fundraising, emboldened openly gay political candidates across several states, and within a year helped push the American Psychiatric Association to stop classifying homosexuality as a mental illness — a cascade none of it linearly planned in advance. The same pattern appears at the individual and organizational level: Michael Phelps's coach Bob Bowman built races around a sequence of tiny, engineered victories (each visualized step of the pre-race routine going exactly as planned) that made the actual race feel like just one more win in an unbroken streak; and after a young Alcoa worker was killed in an accident, Paul O'Neill turned the resulting short-term drop in the injury rate into a rallying point ("we should celebrate because we are saving lives"), using that small win to license a flood of unrelated suggestions from the shop floor, including one that doubled profits on aluminum siding.

Why It Matters

Small wins solve the problem that "big enough to matter" goals are often too large to attempt directly, whether that goal is a social movement, an org-wide culture change, or an individual habit overhaul that keeps failing when attempted all at once. The concept gives a design principle for sequencing change: don't try to prove the whole thesis at once — find and bank the smallest available proof that the pattern can shift, then let the psychological and structural aftereffects of that proof do compounding work. It's a domain-neutral answer to "where do you get momentum before you have any."