A settlement, not a verdict
On August 3, 2026, the Netherlands Authority for Consumers and Markets (ACM) published a draft decision proposing to make binding a set of commitments offered by bol.com, the country's largest online marketplace. The commitments follow an investigation the ACM opened on February 14, 2024, into whether bol favored its own retail offers and those of a favored subset of sellers in how products are ranked, and whether it used data collected from third-party sellers to its own advantage as a competing seller on the same platform. The draft decision is open for public consultation through September 13, 2026, after which the ACM will weigh stakeholder input before finalizing it (ACM, August 3, 2026).
What makes this case instructive is the procedural choice, not just the substance. The ACM is not issuing an infringement decision or a fine. Bol's own account of the settlement states plainly that the ACM did not establish any violation of law. Instead, the case will close through a commitments mechanism under the Mededingingswet (Dutch Competition Act), under which a company's voluntary promises are made legally binding in exchange for the regulator dropping further pursuit of the specific conduct covered. That is a materially different outcome from ACM's 2021 order against Apple over App Store payment terms for dating apps, which involved a formal finding and a penalty structure worth up to €50 million if Apple failed to comply within two months (ACM, 2021).
What bol.com actually agreed to
The commitments cluster around three areas. First, bol will change how its buy-box — the default "koopblok" offer shown to consumers — determines which seller's listing gets top placement, and will share more ranking-relevant data with sellers so they can understand why they did or didn't win the box. Second, bol commits not to use data it collects as the platform operator to advantage its own retail arm when it competes against the same third-party sellers whose data it holds. Third, bol will speed up complaint handling and dispute resolution for business sellers, clarify its terms and conditions, and make its Select Deals promotional program available to sellers on the same terms it applies to itself (ACM draft decision, August 2026).
The case for a formal ruling
There is a genuine argument that the ACM let bol off easy. Self-preferencing by a dominant marketplace operator against the very sellers who depend on it for market access is precisely the kind of structural conflict of interest that competition law exists to police, and a commitments decision — unlike an abuse finding — creates no legal precedent, no fine, and no public finding of wrongdoing that future claimants or regulators elsewhere could build on. Sellers who felt harmed by years of allegedly skewed rankings get no compensation and no acknowledgment that the conduct was unlawful. Critics of platform self-dealing, in the EU and beyond, have long argued that commitments let large platforms treat fines as a cost of doing business avoided altogether, since even a favorable settlement imposes no monetary consequence for past conduct.
Why the commitments route is still the better call here
That argument is strongest when a dominant firm has clearly broken the law and delay only compounds harm. It is weaker here. The ACM itself has not found bol dominant or abusive — its own materials frame the case as "concerns about the functioning of the platform," not an established infringement. Under Article 49a of the Mededingingswet, the ACM may only accept commitments where doing so is more appropriate than pursuing a fine or compliance order, which requires the authority to judge that a faster, verifiable fix serves competition better than years of contested litigation over dominance and causation. Litigating a full self-preferencing case — proving market dominance, defining the relevant market against Amazon, Coolblue, and a fast-growing wave of Temu and Shein-driven marketplace competition, then proving intent and effect — typically takes years and produces a remedy that lands well after the competitive harm has already occurred. Sellers get functioning fixes to ranking transparency and data use within months rather than a ruling half a decade from now.
This also isn't happening in a legal vacuum. The EU's Platform-to-Business Regulation, in force since 2020, already requires online marketplaces to disclose the main parameters determining ranking and to operate an internal complaint-handling system for business users (Regulation (EU) 2019/1150). Bol's commitments substantially track those existing EU obligations rather than inventing novel Dutch-only rules — which is exactly the kind of proportionate, harmonized enforcement regulators should favor over parallel national rulebooks that fragment compliance across the single market.
The real test is verification
The risk with any commitments decision is enforceability drift — promises that look strong on paper but go unmonitored once the headlines fade. The ACM's decision will only be as good as its ability to audit bol's buy-box algorithm and dispute-handling metrics after the fact, not just at the moment of signing. Dutch and EU regulators should treat this case as a template: use commitments where dominance is contested and speed matters, but pair them with real technical audit rights, not just self-reported compliance. Sellers and the public consultation closing September 13 are the last checkpoint before this becomes final.