Despite what we learn in law school about the “meeting of the minds,” most contracts are merely boilerplate—take-it-or-leave-it propositions. Negotiation is nonexistent; we rely on our collective market power as consumers to regulate contracts’ content. But boilerplate imposes certain information costs because it often arrives late in the transaction and is hard to understand. If those costs get too high, then the market mechanism fails. So how high are boilerplate’s information costs? A few studies have attempted to measure them, but they all use a “horizontal” approach—i.e., they sample a single stratum of boilerplate and assume that it represents the whole transaction. Yet real-world transactions often involve multiple layers of contracts, each with its own information costs. What is needed, then, is a “vertical” analysis, a study that examines fewer contracts of any one kind but tracks all the contracts the consumer encounters, soup to nuts. This Article presents the first vertical study of boilerplate. It casts serious doubt on the market mechanism and shows that existing scholarship fails to appreciate the full scale of the information cost problem. It then offers two regulatory solutions. The first works within contract law’s unconscionability doctrine, tweaking what the parties need to prove and who bears the burden of proving it. The second, more radical solution involves forcing both sellers and consumers to confront and minimize boilerplate’s information costs—an approach I call “forced salience.” In the end, the boilerplate experience is as deep as it is wide. Our empirical work should reflect that fact, and our policy proposals should too.