Abstract The European Commission’s new horizontal guidelines create scope for sustainability agreements and, to calculate efficiencies, suggest the estimation of consumers’ willingness to pay (WTP) for so-called “nonuse value.” We contrast estimates obtained from a hypothetical choice experiment conducted for the Netherlands’ competition authority (the “Chicken of Tomorrow” case) with estimates based on homescan data documenting consumers’ real choices between more or less animal-friendly meat, ensuring that the respective alternatives are made highly comparable. Still, the estimated WTP for the animal-friendly alternative is several times higher in the choice experiment. Given observed prices and the WTP estimated in the choice experiment, the animal-friendly alternative should have already crowded out the conventional variant. We use our findings to inform about several pitfalls both when using experimental data and when using data from true purchases.