MI3: Today, we have the pleasure of speaking with two of the authors of a study on crowdsourcing wine quality assessments, Orestis Kopsacheilis, a postdoctoral researcher at the Economics and Policy Department of the Technical University of Munich School of Management, and Ophelia Deroy, one of the principal investigators of MI3. Thank you for joining us! Let’s start with a more philosophical question. What is it about the evaluation of taste that interests you, and why is this a central focus of your research?
Ophelia: That’s a wonderful question. Evaluations of taste are deeply significant, not just in the realm of fine art like music, but also in our everyday experiences with food, wine, perfumes, textiles and all sorts of other sensory pleasures. Taste is about standards of quality, which can vary widely across people, cultures and contexts. But taste is also about discrimination—being able to distinguish between different qualities.
At least, standards and quality is one entry into the values of taste—as distinct from economic value for instance. When you listen to a piece of music you may ask, “Is that a good interpretation? What were its qualities?” It is not just about judging whether it was fast, or loud—as a matter of quantities only. What is fast enough? Was it better than the previous piece? Same for wine: is it good? Is it balanced, or too strong?
It’s something that philosophers like David Hume wrote about extensively. Interestingly, Hume himself started his exploration of taste with an analogy about wine, which makes our study feel even more relevant.
MI3: Fascinating. Can you tell us more about what inspired this particular study?
Orestis: The idea came from noticing how people traditionally rely on either the price of a bottle or expert reviews to judge wine quality. But both of these have their flaws. Price doesn’t always reflect quality accurately, and expert reviews, while insightful, don’t always align with what the average person enjoys. With platforms like Vivino allowing anyone to rate and review wines, we saw a chance to investigate how this ‘wisdom of the crowd’ stacks up against traditional expert assessments. It’s a perfect case of personal taste meeting collective judgment.
MI3: That’s really interesting. So is your research about democratizing wine recommendations?
Ophelia: That is one way of seeing it. The better way perhaps is to check that there is some information that one gets about wine quality from non-expert reviews. This is particularly important in a time when people are turning to their peers—not just experts—for advice and recommendations. In our study, we analyzed over 200,000 reviews of wines from Bordeaux from Vivino users to understand how these crowdsourced ratings align with or diverge from expert opinions.
MI3: And what did you find? How do the opinions of the crowd compare to those of the experts?
Orestis: Our analysis showed that while there is a significant overlap between expert scores and crowdsourced ratings, the correlation isn’t perfect. For example, Vivino users tend to prefer wines when those are closer to their peak drinking age, while experts might rate a younger wine highly based on its future potential. This highlights a key difference: experts often evaluate wines with an eye toward their aging potential, while everyday consumers focus more on immediate enjoyment.
MI3: So it sounds like there’s value in listening to both sources—experts for their deep knowledge and predictions, and the crowd for their current, real-time experiences.
Ophelia: Absolutely. It’s also important to note that the users on Vivino aren’t true novices. Much like in the famous Galton experiment on the wisdom of crowds, where participants betting on the weight of ox were not entirely inexperienced with livestock, Vivino users often have a certain level of experience with wine. They may not be professional critics, but they’ve likely tried enough wines to develop a sense of what they like and can recognize quality to some extent. This brings us to an important conceptual point: expertise isn’t a binary state. It exists on a continuum. The crowd in Vivino represents a range of experiences and knowledge, which together form a robust, collective judgment.
MI3: That’s a great point. How does this idea influence how we should view the role of crowdsourced reviews?
Orestis: When we consider expertise as a continuum, it helps us appreciate the value that these crowdsourced reviews offer. The collective opinions of a large group of people with varying degrees of experience can often allow us to tap into a diverse range of palates and perspectives, which can be incredibly useful for consumers trying to navigate the vast world of wine.
MI3: In your paper, you also discuss the impact of factors like weather conditions on wine ratings. Given the increasing concern around climate change, can you explain why these insights are particularly relevant today?
Orestis: The motivation to test the responsiveness of wine ratings against weather conditions stems from our goal to test the validity of such ratings against a set of objective criteria. Even though the subjectivity of tastes cannot be overstated, there is a wide consensus among wine makers regarding the optimal weather conditions that give rise to the best quality grapes for wine production.
The conditions during the growing season—such as temperature and rainfall—affect the grapes’ development and, consequently, the wine’s flavor and structure. To the extent that judgments about quality contain both objective and subjective components, we expected that such judgments should be able to mirror these yearly weather changes. Indeed, we found that even non-expert reviewers on Vivino are sensitive to these factors, as their ratings often reflect the quality markers associated with favorable weather conditions.
Ophelia: And this is particularly important now because climate change is altering these conditions. Warmer temperatures, for instance, can accelerate grape ripening, leading to wines that might be less balanced or have different flavor profiles than in the past. In fact, we were able to already detect some detrimental aspects of global warming on wine ratings, when high temperatures extend deep into the growing season.
Our findings suggest that as climate change continues to impact vineyards globally, the collective insights from a diverse group of wine enthusiasts could become an even more valuable resource for consumers and beyond.
MI3: How do you see the role of crowdsourced reviews evolving in the future?
Ophelia: I believe crowdsourced reviews will continue to grow in importance, not just for wine, but across many industries. Just like social interactions, recommendations increasingly move online. So the wisdom of the crowd plays a bigger role in shaping our decisions. Many people like to issue red flags—as if people would become more conformist, or stay in their bubbles. Crowds can be smart too, and they can also complement the experts. This is not a replacement, note: we were looking for overlaps between experts and Vivino tasters.
MI3: This has been a really enlightening conversation. Before we wrap up, is there anything else you’d like to add?
Ophelia: Just that I hope our research encourages people to explore their tastes collaboratively.
MI3: Thank you so much, and it is clear that the future of wine tasting is more collective than ever before. Cheers to that!
Ophelia: Cheers! It’s been a pleasure talking with you.
Orestis: Thanks and same here!
Check out the original study and the profiles of Orestis Kopsacheilis and Ophelia Deroy.
References:
Kopsacheilis, O., and Pipergias Analytis, P., Karthikeya, K., Herzog, S.M., Bahrami, B., and Deroy, O. Crowdsourcing the assessment of wine quality: Vivino ratings, professional critics and the weather. https://ssrn.com/abstract=4355578