Distribution advantage

Distribution gets ahead of product competition. Customers compare products only when the products are in front of them at the same moment. A default, a nearby store or an existing bottler shrinks the field before the comparison starts. The court cited a 2017 Google document showing that over 60% of Google’s queries flowed through defaults: 60% of iOS queries through Safari and 80% of Android queries through defaults secured by its distribution deals. Going around the default is hard. Google received only 7.6% of the queries on Apple devices through Chrome that users downloaded themselves.

The route also decides who keeps the margin. Coca-Cola’s history shows this most clearly. In 1899 Benjamin Thomas and Joseph Whitehead secured from Asa Candler the exclusive rights to bottle Coca-Cola across nearly the whole United States. They then found they could not raise enough capital to build plants nationwide, so they contracted local operators to build them in defined territories. By 1920 more than 1,200 bottling operations existed. The company that owned the formula never had to fund most of the trucks and glass. Its 2025 annual report still describes the split: Coca-Cola sells concentrates and syrups to bottling partners, who mix, package and deliver them, and its finished-product operations produce higher revenue but lower gross margins than the concentrate business. In 2024 it refranchised its own bottling operations in parts of India, Bangladesh and the Philippines, moving the capital-heavy link back to partners.

Owning the route pays when density lowers the cost of every sale. Dollar General reported 20,893 stores at the end of fiscal 2025, with about 75% of the U.S. population within five miles of one and about 80% of its stores in towns of 20,000 people or fewer. Its filing attributes its low prices to a small-store format, a limited assortment and a low-cost operating approach. A rival could copy any one of those, but matching the footprint means building thousands of small stores to match the coverage. Consumables were 82% of fiscal 2025 net sales of $42.7 billion, which suggests the route is valuable because customers use it for weekly purchases.

The practical test has three questions. Who controls the route, and can they redirect it? What does each customer reached through it cost compared with a rival’s route? And if a rival offered more money, a better product or a legal challenge, what would the route’s owner do? Google’s payments are large because Apple, as the route’s owner, can weigh Google’s offer against Microsoft’s every time the contract comes up.

A route you do not control can be redirected by its owner. In its 2012 annual report Zynga said it derived 86% of its revenue and 81% of its bookings through Facebook and acquired substantially all of its players there. Facebook kept 30% of what players spent. In December 2012 Facebook changed its terms so that apps on its site could not promote or link to game sites other than Facebook. In 2013 Zynga’s daily active users fell from 63 million to 37 million, revenue fell from $1.28 billion to $873 million and bookings fell 38%. Zynga’s filing attributes the decline to weakness in existing games and a lack of new hits, not to Facebook alone. That is the point: the route was rented, and when the games weakened there was no second route to fall back on.

A good route does not rescue a product customers do not want. In October 2018, announcing the consumer service’s closure, Google wrote that Google+ had not achieved broad consumer or developer adoption and that 90% of user sessions lasted less than five seconds. Reach delivered users to the product; those figures do not show why users left, but they do show that reach had not produced sustained use.

Distribution can also be litigated away. Judge Mehta found that Google had violated Section 2 of the Sherman Act, and his remedies barred exclusive deals and ordered Google to share some search index and user-interaction data with rivals. The court stopped short of banning payments: paid defaults remain allowed if they are not exclusive and run for a year or less. A company that buys its route should expect the terms to be reopened.

Managers overstate the advantage when they measure reach rather than use. Store counts, install counts and partner logos show that a route exists. They do not show what customers do once they arrive.