Have you ever looked at a perfume box and thought its color really suited the fragrance? Does a better-fitting combination then make you more inclined to buy that perfume? These may sound similar, but they are separate questions.
In 2015, Hendrik N. J. Schifferstein and Bryan F. Howell investigated them in sequence by combining commercial fragrances with experimental boxes. Different box colors were rated as appropriate for different fragrances, but better color matches did not produce higher purchase intentions. The findings highlight the importance of distinguishing what fits, what we like, and what we want to buy.
- First, could they create boxes that suited the fragrances?
- Color fit differed, but not in the same way for every fragrance
- Comparing the same fragrance with matching and less-matching boxes
- When the recipient changes, personal liking explains less
- Why the authors did not conclude that color matching is unnecessary
- Summary
- References
First, could they create boxes that suited the fragrances?
If color can communicate a fragrance’s character, it can give people clues to the contents before they smell it. Correspondence between visual and olfactory impressions might also make a product more appealing overall. The authors set out to test these possibilities using fragrance packaging.
They used the eaux de toilette Kouros, DKNY, Paris, and Miss Dior original, and the eau de parfum Wish. Fragrances were applied to blotters and smelled from containers. Participants were not simply shown the retail bottles or original product boxes.
Boxes were created for the research. Their shape, pattern layout, and common non-brand-identifying wording were held constant while colors varied. Each box used three colors selected by a designer with reference to odor–color fit ratings from earlier research. This allowed the researchers to test whether judgments of individual colors that suited a fragrance carried over to boxes containing multiple colors.
Color fit differed, but not in the same way for every fragrance
Experiment 1A involved 66 Dutch women, and Experiment 1B involved 80 American women, recruited from university students and staff. In each experiment, participants rated how well five fragrances matched five boxes.
Results differed by fragrance. In the American experiment, for example, Kouros was rated as fitting green and blue boxes, while Wish and Paris fitted red boxes better. Miss Dior, however, showed no significant differences between boxes in either experiment. Differences between box colors for DKNY observed in the Netherlands were not significant in the United States.
Ratings of the whole box’s fit with a fragrance could largely be explained by how well its largest-area color matched that fragrance. However, the main color not only covered the largest area but also had high fit ratings, and the secondary colors were chosen for harmony. The findings do not establish that only the main color matters in every color scheme.
Thus, although fragrance–box color correspondences may be shared to some extent, they do not identify one correct color for each perfume. Given broadly overlapping patterns in the two countries, the authors next asked whether matching combinations translated into willingness to buy.
Comparing the same fragrance with matching and less-matching boxes
Experiment 2, examining purchase intention, involved 109 American women. It used three fragrances, Kouros, DKNY, and Wish, and three boxes, green, yellow, and red. Based on the earlier experiments, Kouros–green, DKNY–yellow, and Wish–red were designated matching combinations.
There were nine possible combinations, but not everyone evaluated all nine. Each participant evaluated three combinations, assigned so that neither a fragrance nor a box was repeated for the same person. Participants rated fragrance liking, box liking, and their fit, as well as willingness to buy for themselves, a male friend or relative, and a female friend or relative.
Fit ratings again varied with the fragrance–box combination. However, a matching combination did not significantly increase purchase intention for any of the three purposes. Feeling that a combination fitted was demonstrated, but a corresponding increase in wanting to buy was not.
Purchase intention was mainly associated with liking the fragrance itself. For personal use and gifts for men, liking the box was also associated, though less strongly. The important distinction is that liking a box and thinking it fits a fragrance are not the same judgment.
When the recipient changes, personal liking explains less
For gifts to women, participants’ own liking was less strongly related to purchase intention, and purchase intentions were generally lower. To help interpret this finding, the authors interviewed 12 women from the target population about their thinking when buying perfume.
The interviews revealed views of perfume as personal and connected with identity, and approaches that distinguished one’s own preferences from those of the recipient. Some women explained that, when choosing gifts for men, they wanted fragrances they themselves liked because they would also smell them when spending time together.
These interviews supplement interpretation of the experimental differences. Twelve women’s accounts do not determine how women in general choose gifts. They nevertheless suggest that interpreting gift purchase intentions requires considering not only fragrance–box fit but who the recipient is and what perfume means to that person.
Why the authors did not conclude that color matching is unnecessary
The authors argued that it would be premature to conclude that appropriate packaging colors are unimportant merely because no effect on purchase intention was found. A box or bottle first encountered in a store can create expectations about the fragrance and help people find the kind they want. They considered that a consistent impression conveyed through color, shape, and other features might have a role.
This experiment also asked people how much they liked fragrances they actually smelled. The authors suggested that this might have heightened attention to the scent itself. The same weighting has not been established for choices made while viewing a product as a whole. These are the authors’ interpretations; the experiment did not directly demonstrate how packaging works in stores.
There are also points to consider about the strength of the findings. In Experiment 2, the three responses from each participant were analyzed as though they came from different people. Because possible similarity among one person’s responses was not adequately incorporated, statistical judgments warrant caution. The outcome was purchase intention, not actual buying; participants were university-affiliated women, and the boxes were experimental. A nonsignificant result cannot be interpreted as proof that color matching has zero effect in every setting.
Summary
The study examined choosing box colors that suit a fragrance and making people want to buy that fragrance as two separate stages. Differences in fit appeared in ratings, but an associated increase in purchase intention was not found. Liking for the fragrance and box separately was related to purchase intention.
One lesson is that packaging goals need to be distinguished and tested separately: helping people imagine the contents, making the box appealing in itself, or encouraging those who have smelled the fragrance to choose it. The same color match need not provide the same answer to each question. Perceived fragrance fit is one clue to understanding the path toward purchase.
References
Authors: Hendrik N. J. Schifferstein, Bryan F. Howell
Article: Using color–odor correspondences for fragrance packaging design
Journal: Food Quality and Preference 46, 17–25 (2015)
DOI: 10.1016/j.foodqual.2015.06.012
Full original article: Author-posted full text on ResearchGate
Rights: © 2015 Elsevier Ltd. All rights reserved. The author-posted version does not display a Creative Commons license.
