The Problem People Keep Running Into

Every day, millions of people click "I Agree" on documents they have never read. A 2008 study by Lorrie Faith Cranor and Aleecia McDonald estimated that actually reading every privacy policy a typical American encounters in a year would take roughly 76 work days. That figure has only grown since then, as the average Terms of Service document now runs between 5,000 and 10,000 words — longer than many short novels. Spotify's terms exceed 10,000 words. Apple's iTunes agreement has, at various points, been longer than Shakespeare's Macbeth. These are not edge cases; they are the norm.

The core problem is a structural mismatch between the legal form and the practical context. A Terms of Service agreement is a binding contract. Courts in the United States and most of Europe have consistently upheld "clickwrap" agreements — those where a user clicks "Accept" — as legally enforceable, even when it is demonstrably impossible that the user read them. This means the document serves a legal purpose (establishing enforceability) while being designed in a way that defeats its stated communicative purpose (informing the user of their rights and obligations). The gap between those two functions is where the problem lives.

What makes this more than an inconvenience is the content buried inside. ToS documents routinely contain clauses waiving class-action rights, authorizing data sharing with third-party advertisers, granting platforms broad licenses to user-generated content, and limiting liability for service failures. Users who don't read these terms don't know they've surrendered these rights — but legally, they have. The document is not decorative. It has real consequences that surface at the worst moments: after a data breach, a billing dispute, or a sudden account termination.

How Modern Systems Created This

Legal liability expands documents over time. Every lawsuit a platform faces, and every regulatory action it navigates, tends to produce a new clause. Corporate lawyers add language to close the specific gap that exposed the company to risk. Over years, this produces a document that is less a coherent user agreement and more a sedimentary record of every legal threat the company has ever encountered. There is no corresponding pressure to remove old clauses once added, because deletion creates new ambiguity. The result is relentless accumulation with no natural ceiling.

Clickwrap doctrine removes the incentive to be clear. Courts established in cases like Specht v. Netscape (2002) and Meyer v. Uber (2017) that the key question is whether the user had "reasonable notice" that terms existed and an opportunity to review them — not whether they actually understood them. This is a low bar. Once a company posts a link and a checkbox, the legal obligation is essentially met. There is no judicial reward for plain language, no penalty for complexity. The incentive to invest in readable prose simply does not exist within the current legal framework.

Platforms operate under multiple, overlapping regulatory regimes. A global platform like Google or Meta must simultaneously satisfy the requirements of the EU's GDPR, California's CCPA, Brazil's LGPD, and dozens of other frameworks. Each regime has specific disclosure requirements — particular phrases, specific opt-out mechanisms, mandated notices — that must appear verbatim. These requirements don't coordinate with each other. The result is a document that reads like a legal patchwork, because it literally is one: different sections written to satisfy different regulators in different jurisdictions, stitched together into a single document served to every user regardless of location.

Contracts of adhesion shift all drafting power to one side. A standard ToS is a "take-it-or-leave-it" contract, legally known as a contract of adhesion. The user has no negotiating power and no ability to modify terms. This asymmetry means the drafting party — the platform — has every incentive to write terms that favor itself maximally, since the other party cannot push back. Broad data licensing language, mandatory arbitration clauses, and unilateral modification rights (meaning the company can change the terms at any time, often with minimal notice) are all standard features that a negotiating party would never accept but a passive clicker cannot reject.

Why It Keeps Getting Worse

The feedback loop sustaining this system is self-reinforcing. Because users don't read terms, they don't complain about specific clauses. Because they don't complain, legislators face limited pressure to mandate reform. Because reform is limited, platforms have no external reason to change their practices. Meanwhile, the documents keep growing: GDPR alone, which came into force in 2018, caused a measurable spike in the average length of privacy policies across major platforms, as companies added required disclosures without removing existing language. More regulation, paradoxically, can produce longer and harder-to-read documents without producing more informed users.

Market competition does not solve this problem the way it solves most product quality issues. In a normal market, if one product is worse than a competitor's, users switch. But ToS complexity is not a salient feature at the point of sign-up. Users choose services based on features, price, and network effects — not on the readability of legal documents they intend to skip. This means there is no competitive disadvantage to having an unreadable ToS, and no competitive advantage to investing in a readable one. A few companies, like Basecamp, have experimented with plain-language terms, but these remain novelties rather than industry standards, precisely because the market does not reward them.

How People Cope Today

The most practical tool currently available is the browser extension ToS;DR (Terms of Service; Didn't Read), a crowd-sourced project that grades major services from A to E based on the user-friendliness of their terms and highlights specific clauses of concern. It doesn't replace reading, but it surfaces the highest-stakes issues — mandatory arbitration, data sale to third parties, right to delete your account — in seconds. Similarly, services like Privacy Spy and summaries published by the Electronic Frontier Foundation offer plain-language breakdowns of major platforms' policies. These tools work by doing the reading once, at scale, and distributing the result.

For higher-stakes agreements — employment platforms, financial services, healthcare apps — the more reliable approach is targeted reading: skip to the arbitration clause, the data-sharing section, and the termination and account-deletion provisions. These are the sections most likely to affect you adversely and least likely to be summarized anywhere. Searching the document for keywords like "arbitrate," "sell," "license," "modify," and "terminate" takes under five minutes and covers the majority of consequential risk.

The broader pattern here is one that recurs across many modern systems: a mechanism designed to perform a function (informed consent) has been optimized instead for a proxy of that function (legal enforceability). Once the proxy becomes the target, the original goal quietly disappears. Terms of Service documents are not broken — they work exactly as the incentive structure demands. Understanding that distinction is the beginning of navigating them more deliberately, and the foundation for any serious argument that the structure itself needs to change.