Sludge
Friction that serves the institution and costs the user is sludge, not protection.
sludge: Friction added deliberately to stop someone doing a thing they are entitled to do. is excessive or unjustified friction that makes it harder for people to get what they are entitled to or to leave something they no longer want. Thaler coined the term as the mirror of a nudge: A change to how choices are presented that shifts behaviour without removing any option.. The defining question is not how many steps there are but who benefits from each one.
Where the steps go
How it shows up in software
Sludge in software looks like cancellation that requires a phone call, data export buried three levels deep, a refund form that demands an order number the receipt does not show, and a consent banner where accepting takes one click and refusing takes six. It is rarely designed maliciously in one sitting. It accumulates when every team adds one step to protect its own metric.
Using it well
- Run a sludge audit: time yourself cancelling, exporting data, requesting a refund and closing the account, and put the numbers in front of leadership.
- Apply the symmetry test. Whatever it takes to start should be roughly what it takes to stop, in the same channel.
- Give each required step an owner who can state the user benefit. Steps with no answer get deleted.
- Track burden as a metric, such as median time to cancel, and review it beside conversion so the two are traded off openly.
Where it turns manipulative
- Sludge used as retention is a business model built on people giving up. Revenue from users who wanted to leave and could not is borrowed, and it comes back as churn: The rate at which customers stop paying or stop using the product., chargebacks and regulatory attention.
- Hiding entitlements behind paperwork, such as rebates or refunds designed to be abandoned, is deception even when every step is technically disclosed.
- Regulators now treat this as conduct, not design. The EU Digital Services Act restricts interfaces that make choices harder than necessary, and several jurisdictions require cancellation to be as easy as signup.
Where you have seen it
Cookie consent banners
Accept all is a single button; refusing means opening preferences and toggling vendor lists. Nouwens et al. (2020) found most consent interfaces they crawled fell short of the legal standard.
Subscription cancellation by phone
Signup is a web form, cancellation is a call during business hours to an agent with a retention script. The asymmetry is the whole mechanism.
Apple App Store subscription management
The counterexample. One settings screen lists every subscription with a cancel control, which removes the asymmetry at the platform level.
What the research says
- Thaler, 2018 (Science)Well evidenced
A short editorial naming sludge and arguing that behavioural science had been better at adding helpful nudges than at removing harmful friction.
An editorial, not a study. Its contribution is the concept and the diagnostic question, not new data.
- Sunstein, 2019 (Sludge Audits, Behavioural Public Policy)Mixed evidence
Proposes a systematic audit of the time, paperwork and waiting an institution imposes, and argues these costs fall hardest on the people least able to absorb them.
- Herd and Moynihan, 2018 (Administrative Burden)Well evidenced
Documents how learning costs, compliance costs and psychological costs in US benefit programmes cause large numbers of eligible people to go without benefits.
Careful public administration research rather than lab work. The distributional point is the important one: burden is regressive.
Grades are a judgement about the evidence, not about how useful the idea is. Plenty of contested effects are still worth knowing, as long as you do not cite them as settled.