Purpose – Cobranding is an often used marketing strategy within the theme park industry and it has existed in one form or another since the 1930s. Notwithstanding the growing interest for cobranding in the theme park industry academic research in a theme park context has not been found yet. Empirical research on cobranding is limited to a relatively few studies that have typically examined product concepts or fictitious products rather than real instances of cobranding. This article aims to present results of an experiment on the effects of cobranding from a reallife theme park perspective. Design/methodology/approach – The article is based on a classical field experiment in which the IBRAmethod of measuring brand associations was used. The IBRAmethod does not influence the brand associations like many other research techniques do (by giving certain cues). It is an unaided, unbiasing research technique. The objective of the study is to investigate whether the relationship between theme park Efteling and WWF, resulting in the cobranded attraction PandaVision, could have a negative effect on the strong brand associations of theme park Efteling. Findings – Through the field experiment an insight has been given into the possible effects a respondent's perceived brand fit within a cobranding situation can have on the average evaluation of core associations of one of the constituent brands. Even strong brands (Efteling is the strongest brand in The Netherlands) can be harmed by a wrong cobrand strategy. Results also showed that the brand fit manipulation has resulted in a more negative image of Efteling without affecting the evaluation of the cobranded attraction PandaVision. Only measuring whether guests like or dislike your attractions is thus not sufficient. Research limitations/implications – This research is presented as a preliminarily study and the results should be interpreted with caution. The sample size was limited to 70 respondents and the experimental design with only students may not necessarily represent the typical visitor to the Efteling. Because of the crude manipulation of the treatment it is unclear what precisely caused the established effect. Is the effect caused by the degree of elaboration (meaning, because the respondent is triggered to think deeply about the matter at hand) or by the substantive guidance? Supplementary research with several experimental groups is needed to answer this question. Practical implications – Theme parks should be aware of the dangers of cobranding. Pairing with a wrong partner can damage the brand; negative spillover effects, erosion, brand dilution and even negative bottom line effects for the participating brands are possible. If the results occur for strong brands, weaker brands should be even more aware of the dangers. Originality/value – This article presents the first application of the effects of cobranding in a specific theme park setting. It is also the first article to use the unbiased IBRAmethod for measuring brand associations of a cobrand strategy. Negative effects of cobranding for strong brands in a reallife situation were never reported before.