Hypernudge
Data-driven, personalized, real-time nudging powered by algorithms and continuous behavioral feedback.
What it means
A hypernudge is the algorithmic intensification of nudging, in which big data and machine learning are used to continuously profile individuals and dynamically reconfigure their choice environment in real time — personalized, adaptive, and operating at scale across digital platforms. Where a classic nudge is a fixed feature of the environment, a hypernudge updates moment to moment, learning what moves each person and adjusting defaults, framing, ranking, and prompts accordingly. This makes it far more potent and far harder to detect or resist, raising acute concerns about autonomy, transparency, and manipulation, since the influence is opaque, individualized, and relentless. The term, coined by Yeung, situates it within data-governance and surveillance debates as much as behavioral ethics. It matters because the line between helpful personalization and covert control narrows as systems get better at predicting and shaping behavior — the hypernudge is where choice architecture meets the attention economy.
Examples
A platform that constantly reorders your feed and tunes notifications to whatever currently maximizes your engagement is hypernudging — nudging that learns and adapts to you in real time.
A shopping app learns you buy when the counter says 'two left', so it shows you scarcity banners it never shows your neighbour — and quotes you a different price.
A fitness app quietly tests dozens of wordings on you, keeps whichever gets you off the sofa, and retires the rest — a nudge that rewrites itself every week.
First described in Karen Yeung (2017).