Abstract
Introduction:
Health warning labels (HWLs) are an important way to educate the public about the dangers of tobacco products. Tobacco companies conducted research to understand how pack colors affect consumers’ perceptions of the products and make packages and their labeling more visually prominent.
Methods:
We analyzed previously secret tobacco industry documents concerning the tobacco industry’s internal research on how cigarette package colors and design influence the visual prominence of packages and consumers’ perceptions of the harmfulness of the products.
Results:
The companies found that black is visually prominent, placing dark pack elements on a contrasting light background makes them stand out more, and black text on a white background is more prominent than white text on a black background. Yellow most quickly and effectively seizes and holds consumers’ attention and signals warning or danger, while white connotes health and safety. Using black text on a bright contrasting background color, particularly yellow, attracts consumers’ attention to the message.
Conclusions:
Tobacco industry research on pack color choices that make pack elements more prominent, attract and keep consumers’ attention, and convey danger instead of health should guide governments in specifying requirements for HWLs. These factors suggest that HWLs printed on a yellow background with black lettering and borders would most effectively seize and keep consumers’ attention and signal the danger of cigarettes and other tobacco products.
Implications:
Tobacco companies’ internal research on improving the prominence of pack elements suggests that HWLs using black lettering on a contrasting yellow background would most effectively seize and hold consumers’ attention and signal the danger of cigarettes and other tobacco products.
Introduction
Strong health warning labels (HWLs) on tobacco products are a tool to address the global tobacco epidemic.1 The effectiveness of health warnings depends on their size and design,2,3 and plain packaging may increase their effectiveness.4 The implementation guidelines for the WHO Framework Convention of Tobacco Control recommend that, “Parties should select contrasting colours for the background of the text in order to enhance noticeability and maximize the legibility of text-based elements of health warnings and messages.”5
Tobacco companies have long understood that pack colors influence how consumers perceive the products within these packages and have conducted marketing research since as early as the 1950s to analyze the nature, psychological effects, and impact of color on consumers’ purchasing choices and to analyze how color could be used to design effective packaging.6,7 They use cigarette pack colors to manipulate consumers’ brand choices and perceptions of harm8–14 and to change smokers’ perceptions of the taste and strength of the cigarettes inside the pack without changing the cigarettes themselves.6 The companies conducted extensive research to better understand how changes to the pack colors could effectively create new tobacco products, just as changes to the cigarettes’ ingredients, additives, or flavorings.6 In the course of their research, they tested whether some pack colors are more visually prominent than others, and whether some pack colors can better seize and hold consumers’ attention than others. This research can inform design of effective HWLs.
Methods
Between January 2013 and September 2015, we analyzed previously secret tobacco company documents available at the UCSF Truth (formerly Legacy) Tobacco Industry Documents Library (https://industrydocuments.library.ucsf.edu/tobacco/) on the tobacco industry’s internal research on how cigarette package colors influence the visual prominence of packages. We used standard snowball search techniques,15 beginning with the search terms “(color or colour) and pack*” (225,015 documents), “health warning labels” (2,283 documents), “(health warning labels) and (color or colour) and pack*” (780 documents), “(color or colour) and pack* and ‘visual prominence’” (100 documents), “(color or colour) and pack* and attention” (57,578 documents), and “(color or colour) and pack* and ‘seize attention’” (four documents). Additional documents were found by reviewing documents with adjacent Bates numbers. These searches identified documents detailing the companies’ market research and scientific literature reviews concerning the use of color in package design to understand how to make cigarette packs and their elements more prominent. We reviewed approximately 150 industry documents; this paper is based on 20 of them.16–35
Results
In the 1980s, the tobacco industry became concerned that they would be faced with increasing restrictions on advertising, which would make the cigarette pack itself more important since it would likely be “the last means of communicating with the consumer.”16,17 To meet these challenges, research techniques were developed to ensure that the pack design would be used in the most effective way not only to convey brand imagery but also to create visual impact by achieving a high level of visual prominence.16,18–21 The companies used several techniques to help them understand the relationship between visual prominence and cognitive salience.
A conventional tachistoscope was used to measure the visual prominence of various elements within a cigarette pack by measuring in milliseconds the amount of time it takes for a person to identify particular pack elements. This technique could indicate, within any one pack, how the various color combinations work to make the design elements stand out.16,18–21 The projection tachistoscope was used to measure the standout of various cigarette pack designs in interaction with other packs at the point of sale, as well as to measure the impact of different types of pricing stickers, point of sale add-on material, and the best positioning for the brand logo in posters.18
Eye gaze monitoring was used to track an individual’s viewing behavior over a stimulus field (such as shelving displaying tobacco products at the point of sale) and was used to understand where a consumer directs his attention at the point of sale and how long the consumer visually inspects each discrete point of interest.18 To aid their pack design research, companies introduced high-resolution computer-aided graphics systems in 1985 to allow a large number of variants to be developed and tested.17
The companies’ research showed that some colors (red and black) are more visible and prominent than others (gray), make information more memorable, and appear to advance and make one pack look larger and more visible than neighboring packs22 (Table 1). The companies learned from marketing research covering the use of specific colors to attract buyers that yellow is the first color that the consumer’s eye sees and is processed the fastest, is the most likely color to get quick attention, is seen from a greater distance than any other color, and is retained in the memory longer.22–25,36 Yellow has the highest “retention value” of all colors, which means that “it hits you harder in the eye and you will remember it longer.”36 However, yellow does not make the customer want to buy the product because it is not perceived as pleasant25 and has a low preference rating.36 Although yellow is not well liked, it is easily recalled.36 Yellow increases stress and anxiety,24 and according to industry advisor Carlton Wagner, Director of the Wagner Institute for Color Research, yellow is found in nature in combination with black to notify or warn you of danger (eg, stinging bees are yellow and black).23 Since black is visually dominant, it attracts and keeps consumers’ attention to package elements.21,22,26 The combination of yellow and black add to the retention power of an image.36 For example, although triangles are images that can be easily recalled, a triangle in black or white does not remain in the memory as long as a triangle in black on yellow or black on orange.36
Table 1.
Summary of Industry Research on How Packaging Color Impacts the Visual Effectiveness of Labeling
| Impact | Color | Company | Year |
|---|---|---|---|
| Visual prominence | Black | RJR | 200122 |
| Red | RJR | 200122 | |
| BAT | 198620 | ||
| BAT | 1980s21 | ||
| Seizes attention | Red | RJR | 200122 |
| Yellow | RJR | 200122 | |
| PM | 199525 | ||
| PM | 199023 | ||
| PM | 199024 | ||
| Contrast | Red | BAT | 1980s21 |
| Black on white | RJR | 200122 | |
| BAT | 1980s21 | ||
| BAT | 198620 | ||
| Fast | Yellow | PM | 199023 |
| PM | 199024 | ||
| RJR | 200122 | ||
| Warns of danger | Yellow | PM | 199023 |
Although certain color pack elements (eg, red or black) tend to be prominent, the background on which these elements are placed plays an important role in determining the total effect. Tobacco company tests revealed that placing design elements against a white background, using high color saturation, and creating a good figure-ground contrast are all features that make package design elements visually prominent.20,21 Finding that contrast is the most important factor in determining visual standout, in 1986 BAT researchers concluded, “The rule is very simple, if the element has high colour contrast with its background it will stand out, if it does not then it will not.”20 BAT’s study found that this effect may be accentuated when the dark background on which the object is placed forms part of the pack detail itself.20
Of particular relevance to the design of effective warnings, industry research revealed that dark pack elements on a light background are more visually prominent than light elements on a dark background. BAT used tachistoscopic testing to compare the relative visual prominence of elements on the Muratti Ambassador Multifilter cigarette packs with those on Muratti Ambassador Extra packs. Although the brand name, font style, and color combinations on the two packs were the same, the brand name on the Multifilter pack was light and presented on a dark background, while the brand name on the Extra pack was dark and presented on a light background (Figure 1). The test subjects identified the Extra brand in 7.0ms, one-third faster than the 10.5ms it took to identify the Multifilter brand, confirming that a dark figure presented on a light background stands out more than a light figure on a dark background.20 Likewise, RJR’s research on revised Camel Filters packs reached a similar conclusion: changing the pack from brown with white print to white with brown print resulted in reduced strength perceptions as well as better brand name recognition, and printing the UPC code over a brown background made it less noticeable.27
Figure 1.
BAT tachistoscopic test results on Muratti Ambassador cigarette packs showing that black text on a white background (top) stands out more and is recognized more quickly than white text on a black background (bottom).20
This illustration is from a previously secret tobacco industry document, which is only available as a low quality image. The document is feely available at the UCSF Truth Tobacco Documents Library.
The companies’ research into the emotional impact of colors25,28,29 also revealed that certain colors used on packaging create nonverbal and unconscious impressions about the health dangers associated with smoking. A 1986 BAT research and development report on pack design found that “white is generally held to convey a clean, healthy association.”20 Lighter colors on packs are associated with “light” and “mild” cigarettes, and tobacco companies scientifically adjust the amount of white space or the tone of a particular pack color in order to make consumers perceive that the cigarettes inside the packs are lighter or milder or have lower tar delivery.20,26,27,30 Since the enactment of the Family Smoking Prevention and Tobacco Control Act in the United States which prohibited the use of “light” and “mild” descriptors,37 the companies have replaced those words with light colors that communicate “light” and “mild.”38,39
Discussion
Tobacco company research demonstrating that specific color choices can make pack elements more prominent and seen more immediately, can attract and keep consumers’ attention, can convey danger, and can inform governments in designing required HWLs. Contrast is the most important factor in determining visual standout and dark text and elements on a light background are seen more quickly than light elements on a dark background. Black is visually dominant and attracts and keeps consumers’ attention to package elements,21,22,26 while yellow has the greatest visibility, is the first color seen by consumers, is processed the fastest, and universally signals warning or danger.22–25,36
The tobacco industry’s research is consistent with health research showing that more visually prominent and salient warning labels are more effective.2,3 It is also consistent with industrial safety research on the influence of color on warning label perceptions that found that color enhances the saliency of warnings, increases the readability and level of communicated hazard, draws attention, and influences memory, and recommends that label designers should use background colors that maximize luminance and chromatic contrast with the warning text.40 Studies conducted by social psychologists found that warning label effectiveness is enhanced by using bright colors like orange which attracts consumers’ attention and prompts them to consider the message content and conveys feelings of danger or warning that increase the messages’ impact.41
Both tobacco industry and independent research suggest that black letters on a highly visible, bright, and contrasting background (yellow or orange) would be most effective for presenting the text components of warning labels and would be more visibly prominent and effective than labels that use black lettering on a white background or white lettering on a black background. These findings also support the WHO Guidelines to FCTC Article 115 which recommend using contrasting colors for the background of the text to enhance noticeability and maximize the legibility of text-based elements of health warnings and messages. However, FCTC does not provide any specific recommendations for which colors to use.
Our analysis did not identify any industry research specific to warning label design in the context of total package design either alone or in the context of total package design. Future research is needed to understand how consumer perceptions of HWLs may be altered depending on the extent that the warning label colors contrast with other elements of the pack.
The tobacco companies’ research into the nonverbal impact of colors25,28,29 is also helpful in understanding how warning labels could be designed that create subconscious impressions of the health dangers associated with smoking. Even with labels that carefully adhere to legal restrictions governing words and images, the fast, nonverbal and unconscious response triggered by colors can break through to more powerfully communicate desired statements. All of these factors suggest that HWLs printed on a yellow or orange background with black lettering and borders would most effectively seize and keep consumers’ attention and signal the danger of cigarettes and other tobacco products.
Funding
This work was funded by National Cancer Institute grant CA-87472. The funding agency played no role in the conduct of the research or preparation of the manuscript.
Declaration of Interests
None declared.
References
- 1. World Health Organization. WHO Report on the Global Tobacco Epidemic: The MPOWER Package 2008 http://www.who.int/tobacco/mpower/gtcr_download/en/ Accessed September 22, 2015.
- 2. Hammond D. Health warning messages on tobacco products: a review. Tob Control. 2011;20(5):327–337. [DOI] [PubMed] [Google Scholar]
- 3. Riordan M. Tobacco Warning Labels: Evidence of Effectiveness 2013. http://www.tobaccofreekids.org/research/factsheets/pdf/0325.pdf Accessed September 22, 2015.
- 4. Wakefield M, Coomber K, Zacher M, et al. Australian adult smokers’ responses to plain packaging with larger graphic health warnings 1 year after implementation: results from a national cross-sectional tracking survey. Tob Control. 2015;24(suppl 2):ii17–ii25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. World Health Organization. Guidelines for Implementation of Article 11 of the WHO Framework Convention on Tobacco Control (Packaging and Labelling of Tobacco Products) 2008. http://www.who.int/fctc/guidelines/adopted/article_11/en/ Accessed September 22, 2015.
- 6. Lempert L, Glantz S. Beyond the law: packaging colour research by tobacco companies and implications for regulatory policy. Tob Control. In press. [DOI] [PMC free article] [PubMed]
- 7. Cheskin L. Color Guide for Marketing Media 1954 Brown & Williamson; http://legacy.library.ucsf.edu/tid/yjb14f00 Accessed December 20, 2012. [Google Scholar]
- 8. Wakefield M, Morley C, Horan JK, Cummings KM. The cigarette pack as image: new evidence from tobacco industry documents. Tob Control. 2002;11(suppl 1):I73–I80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wakefield MA, Germain D, Durkin SJ. How does increasingly plainer cigarette packaging influence adult smokers’ perceptions about brand image? An experimental study. Tob Control. 2008;17(6):416–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bansal-Travers M, Hammond D, Smith P, Cummings KM. The impact of cigarette pack design, descriptors, and warning labels on risk perception in the U.S. Am J Prev Med. 2011;40(6):674–682. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Hammond D, Dockrell M, Arnott D, Lee A, McNeill A. Cigarette pack design and perceptions of risk among UK adults and youth. Eur J Public Health. 2009;19(6):631–637. [DOI] [PubMed] [Google Scholar]
- 12. Hammond D, Parkinson C. The impact of cigarette package design on perceptions of risk. J Public Health (Oxf). 2009;31(3):345–353. [DOI] [PubMed] [Google Scholar]
- 13. Mutti S, Hammond D, Borland R, Cummings MK, O’Connor RJ, Fong GT. Beyond light and mild: cigarette brand descriptors and perceptions of risk in the International Tobacco Control (ITC) Four Country Survey. Addiction. 2011;106(6):1166–1175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Yong HH, Borland R, Cummings KM, et al. Impact of the removal of misleading terms on cigarette pack on smokers’ beliefs about ‘light/mild’ cigarettes: cross-country comparisons. Addiction. 2011;106(12):2204–2213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Anderson SJ, McCandless PM, Klausner K, Taketa R, Yerger VB. Tobacco documents research methodology. Tobacco Control. 2011;20(suppl 2):ii8–ii11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. British American Tobacco Company. Product Technology: Brand Development Course 1985. British American Tobacco; https://industrydocuments.library.ucsf.edu/tobacco/docs/gzbb0200 Accessed September 22, 2015. [Google Scholar]
- 17. Sandhu P. A Description of a Computer Aided Graphics System for Pack Design 1985. British American Tobacco; https://industrydocuments.library.ucsf.edu/tobacco/docs/hfbm0205 Accessed September 22, 2015. [Google Scholar]
- 18. British American Tobacco Company. Product Technology Primer 1984. British American Tobacco; https://industrydocuments.library.ucsf.edu/tobacco/docs/xsxw0203 Accessed September 22, 2015. [Google Scholar]
- 19. British American Tobacco Company. Product Technology Seminar: GR&DC 17th-21st October 1983 - Flavour Application Sheet 1983. British American Tobacco; https://industrydocuments.library.ucsf.edu/tobacco/docs/rppm0207 Accessed September 22, 2015. [Google Scholar]
- 20. Miller L. Principles of Measurement of Visual Standout in Pack Design: Report No. RD 2039 1986. British American Tobacco; http://legacy.library.ucsf.edu/tid/ihb34a99 Accessed January 7, 2013. [Google Scholar]
- 21. Product Communication. British American Tobacco; http://legacy.library.ucsf.edu/tid/okz16a99 Accessed January 7, 2013. [Google Scholar]
- 22. Bockweg T, Chapple N, Cornell C, Swaim WF. Color Documentation for Doral Packaging Colors 2001. RJ Reynolds; http://legacy.library.ucsf.edu/tid/hsu14j00 Accessed June 12, 2013. [Google Scholar]
- 23. Color Scheming 1990. Philip Morris; http://legacy.library.ucsf.edu/tid/wkn86c00 Accessed December 17, 2012. [Google Scholar]
- 24. Wagner C. Color Cues 1990. Philip Morris; http://legacy.library.ucsf.edu/tid/xkn86c00 Accessed December 19, 2012. [Google Scholar]
- 25. The Total Package 1995. Philip Morris; http://legacy.library.ucsf.edu/tid/wmd42d00 Accessed December 19, 2012. [Google Scholar]
- 26. Project XG Packaging Briefing Document 1984. RJ Reynolds; http://legacy.library.ucsf.edu/tid/lkc85d00 Accessed January 8, 2013. [Google Scholar]
- 27. Etzel E. Camel Filter Revised Packaging Test: Consumer Research Proposal 1979. RJ Reynolds; http://legacy.library.ucsf.edu/tid/qxb79d00 Accessed December 11, 2012. [Google Scholar]
- 28. Hine T. The Total Package: Seeing and Believing 1995. Lorillard; http://legacy.library.ucsf.edu/tid/uxh54a00 Accessed December 21, 2012. [Google Scholar]
- 29. Brand Marketing Training Module RJ Reynolds; http://legacy.library.ucsf.edu/tid/gqy09d00 Accessed December 21, 2012. [Google Scholar]
- 30. Packaging 1987. RJ Reynolds; http://legacy.library.ucsf.edu/tid/yre05d00 Accessed January 2, 2013. [Google Scholar]
- 31. Sun Research Corporation. Executive Summary. A Qualitative Focus on New Product Development for Marlboro: Marlboro Milds 1998. Philip Morris; http://legacy.library.ucsf.edu/tid/wdy89h00 Accessed September 12, 2014. [Google Scholar]
- 32. Bennette A, Ellis N. Marlboro Milds Research - Final Report 1998. Philip Morris; http://legacy.library.ucsf.edu/tid/feb99h00 Accessed September 12, 2014. [Google Scholar]
- 33. Monieson M. Milds Research Findings 1998. Philip Morris; http://legacy.library.ucsf.edu/tid/xdy89h00 Accessed September 12, 2014. [Google Scholar]
- 34. Cato G. Savings Segment Project for R.J. Reynolds Tobacco Company a Visual Category Investigation First Presentation 1991. RJ Reynolds; http://legacy.library.ucsf.edu/tid/rup03d00 Accessed December 19, 2012. [Google Scholar]
- 35. Viceroy Rich Lights Product, Pack, and Advertising Testing of Blue and Silver Packs 1978. Brown & Williamson; http://legacy.library.ucsf.edu/tid/ews00f00 Accessed January 23, 2013. [Google Scholar]
- 36. Cheskin L. How to Predict What People Will Buy. New York, NY: Liveright Publishing Corporation; 1957. [Google Scholar]
- 37. Family Smoking Prevention and Tobacco Control Act, Pub. L. No. 111-31, Section 911(b)(2)(A)(ii) (2009).
- 38. Anderson SJ, Ling PM, Glantz SA. Implications of the federal court order banning the terms “light” and “mild”: what difference could it make? Tob Control. 2007;16(4):275–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Connolly GN, Alpert HR. Has the tobacco industry evaded the FDA’s ban on ‘Light’ cigarette descriptors? Tob Control. 2014;23(2):140–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Braun C, Kline P, Silver N. The influence of color on warning label perceptions. Int J Ind Ergonomics. 1995;15(3):179–187. [Google Scholar]
- 41. Strahan EJ, White K, Fong GT, Fabrigar LR, Zanna MP, Cameron R. Enhancing the effectiveness of tobacco package warning labels: a social psychological perspective. Tob Control. 2002;11(3):183–190. [DOI] [PMC free article] [PubMed] [Google Scholar]

