Nudge Wiki

Concept

Message Brevity Effect

writing communication brevity

Shorter messages consistently outperform longer ones at driving a reader to actually take the requested action, even though most writers — and most readers, when asked to predict — assume the opposite. Todd Rogers identifies the underlying error: writers imagine successfully transferring everything compelling in their own head into the reader's, but in practice readers are busy, skim or skip anything that looks effortful, and often abandon a message partway through rather than reading it in full. Winston Churchill's 1940 wartime memo "Brevity," demanding his cabinet write shorter reports to save time and "discipline thinking," anticipates the same finding by 80 years.

Editors instinctively add rather than cut, even when cutting is the goal. A University of Virginia study (Gabrielle Adams) asked participants to edit a short article; 83% of participants added words rather than removing any, despite editing normally being framed as an exercise in improving concision.

A controlled field test with school-board members. Rogers tested two versions of an email to 7,000 elected US school-board members, differing only in length: a 127-word version full of deferential, appreciative language, versus a 49-word version that cut straight to the request. 93% of people shown both versions predicted the longer, more deferential email would perform better. It didn't — the 49-word version drove a 4.8% response rate versus 2.7% for the longer one, more than twice as effective (see School-Board Brevity Email Study). Rogers attributes part of the effect to a signaling mechanism: a long message implies a long, effortful task (e.g. a lengthy survey) even when the actual ask is identical, so readers self-select out based on message length alone, not just its content.

The effect replicates outside the lab, at scale, and with editing done by non-experts. A journalist-focused newsletter, edited down from 566 to 275 words by its own staff in 30 minutes (removing lower-priority detail but keeping the same six linked resources), roughly doubled the proportion of the newsletter's 50,000 subscribers who clicked through (see Journalist Newsletter Halving-Words Study). A US political fundraising email sent to 700,000 donors, with entire paragraphs arbitrarily deleted (rather than carefully edited) to roughly halve its length, raised donations 16% — despite readers rating the shortened version as less coherent than the original (see Political Donor Paragraph-Deletion Study). Even a 36-word text message to parents beat an already-short 20-word version by a further 6% in survey response rate (see Virginia Summer Text-Message Brevity Study).

Fewer links work the same way as fewer words. A promotional email for a print magazine at behavioralscientist.org, stripped down to a single link (the one to actually buy the magazine) rather than several, produced a 50% increase in click-throughs and a comparable increase in sales (see Behavioral Scientist Single-Link Email Study).

Fewer requested actions also outperforms more. A climate-change appeal offering 1,500 participants a randomly-chosen shortlist of five suggested behaviors (rather than a full list of 20) produced a higher number of actions actually taken than offering all 20 (see Climate-Change Fewer-Actions Study).

Rogers frames three distinct kinds of "less is more": fewer words (the easiest to apply), fewer ideas (harder — requires judging what to cut), and fewer requests (asking for one thing, not several, since adding a second request measurably reduces the odds either one gets done).

A wider treatment of "easy to read" (Episode 185), opening with the classic illustration of what happens when nobody reads at all. A UK Wi-Fi provider's stunt legally bound 22,000 users to 1,000 hours of community service via a 3,159-word terms-and-conditions clause; only one person noticed (see Purple Wi-Fi Terms-and-Conditions Stunt Case). Eye-tracking shows readers pause and attempt to synthesize meaning at every sentence's period, forcing a re-read when the sentence was too complex to parse in one pass (see Period-Pause Eye-Tracking Study). Complexity has consequences beyond marketing: complex ballot language measurably increases voters skipping a question entirely (see Ballot-Question Complexity Study (2011)), and simpler writing predicts more Facebook likes and TripAdvisor helpful-votes even when the writer's goal isn't pure information transfer (see Facebook and TripAdvisor Simplicity Study). A 2023 study found switching a laundry-detergent ad from technical to plain language raised purchase intent 11.9% with no change in actual content (see Laundry Detergent Technical-Language Study (2023)) — though NIH grant funding data shows a genuine exception, where more complex proposal writing correlates with (not necessarily causes) higher funding rates, a reminder that brevity is a strong default, not a universal law (see NIH Grant-Proposal Complexity Paradox Case).

Beyond word count: perspective, formatting, messenger, and medium. Writing from the reader's likely values rather than the writer's own goal quadrupled Rock the Vote's volunteer sign-ups from an identical ask (see Rock the Vote Reader-Perspective Study). Bold/underlined text reliably draws attention to the formatted content, but at a real cost — it measurably reduces attention to everything else on the page, since readers appear to treat formatting as license to skip the rest (see Bold/Underline Formatting Studies). A letter's apparent sender matters as much as its wording: a tax-credit letter from California's official tax authority outperformed an identical letter from a nonprofit by 3x (see Franchise Tax Board Messenger-Authority Study). And for actions that can't be completed the instant a message is received, physical mail reliably beats email, since a postcard becomes a lingering physical reminder in a way a digital message doesn't (see CalFresh Postcard-vs-Email Study).

Re-cited in the 2024 year-end recap with a large-scale confirmation (Episode 218): a 2016 Boomerang analysis of roughly 40 million emails found the highest response rates for messages between 50 and 125 words, with response rates declining beyond that range — corroborating the school-board study's finding at far greater scale (see Boomerang 2016 Email-Length Meta-Study).

Discussed in

  • Episode 177 — 177-the-overlooked-secret-to-better-writing
  • Episode 185 — 185-how-22-000-were-persuaded-to-clean-sewage
  • Episode 218 — 218-the-best-of-nudge-in-2024 (Boomerang 2016 meta-analysis; school-board study re-cited)
  • Episode 233 — 233-will-tips-from-a-102-year-old-marketing-book-work-in-2025 (re-cited against Claude Hopkins's "more you tell, more you sell" claim; new 776,000-recipient fundraising study)

Related