Launching a line extension (a variant), such as a new flavour or scent, is expensive and risky, with an average of 40% failing to remain on shelf two years after launch (Castellion and Markham, 2013; Victory et al., 2018). In addition to how it is priced, and where it is sold, how effectively the packaging communicates variety is also likely to contribute to a new launch’s chance at survival.
If one brand in a category launches a new flavour and it has a modicum of success, competitor copycats tend to follow. If there is consistency in the visual elements that brands use to signal these varieties, this prototypical design can develop in the minds of category buyers e.g., green = organic. Some prototypes come from the source ingredient (e.g., purple for lavender scent reflecting the flower colour), while others are created by the pioneering brands in the category. For example, the blue and green colour combination for Ranch flavoured dressing in the USA draws from the original Hidden Valley design.
To effectively signal the variety, these prototypical colours must be processed by category buyers. Yet, by conforming to category prototypes, brands run the risk that their own brand colours are negated on pack.
This report provides clarity on the above by addressing the following five questions:
- Do brands within a category use consistent colours on variants?
(e.g., do charcoal toothpastes use the same colours to signal this variant?)
- Do category buyers have consistent expectations of variant colours?
(e.g., how many toothpaste category buyers expect charcoal variants to be a specific colour?)
- Do category buyer expectations match marketing practice?
(e.g., if toothpaste brands consistently use black for charcoal variants, do category buyers also expect charcoal variants to have black on the packaging?)
- Do category buyers have consistent expectations of variant images?
(e.g., how many toothpaste category buyers link an image of charcoal to charcoal variants?)
- How do images perform, relative to colours, as signals of variant type?
(e.g., how does the proportion of category buyers who link an image of charcoal compare to the proportion who link the colour black?)
1. Do brands within a category use consistent colours for the same variants?
To understand the value of prototypical colours to communicate variety on pack, we must first quantify their existence.
Data:
We grouped 576 products belonging to 80 brands into 25 variant types and coded them for their use of colour on-pack. A colour was coded, regardless of its prominence, so long as it appeared on the front of the pack. The products spanned three product categories (chewing gum, fabric conditioner and toothpaste) in the USA.
Method:
We classed a colour as prototypical to a particular variant type if:
- It is present on at least 50% of products of that variant type; and,
- The colour is used by that variant significantly more often than by the rest of the category (i.e., is not just common to the entire category).
Key results:
Colour is often used to signal the type of variant. We found 21 of 25 variant types (84%) have at least one prototypical colour, and this was true for all three categories. On average, 80% of individual products use these prototypical colours. Among variant types that have a prototypical colour, two in three have more than one, which is typically presented as a colour combination (e.g., red and pink for strawberry flavoured gum). The specific colours prototypical to product varieties within each of the categories are presented in Table 1 below.
2. Do category buyers have consistent expectations of variant colour?
While brands may use colour signals on pack, this does not necessarily mean these signals are noticed by category buyers and used to navigate between products in a portfolio. Here we investigate the extent to which consumers hold expectations of packaging colour for specific variant types.
Data:
We surveyed 1853 category buyers across the three categories in the USA.
Method:
Respondents were presented with a swatch of a single colour and provided with a matrix of greyscale, mock-up product images for different variant types for a fake brand. They were then asked to pick which, if any, of the displayed variant(s) was best represented by the colour shown. A ‘none of these’ option was provided to discourage guessing.
Two criteria were used to classify a colour as something that is generally expected by consumers as a signal for a particular variant type:
- The colour is linked to the variant by at least 50% of category buyers; and
- The colour is linked to that variant significantly more often than to other variants.
Key results:
Across three product categories, 14 of 25 variant types have at least one colour that is expected by category buyers. However, consumer expectations vary significantly between categories. Category buyers have an expectation of packaging colour for most fabric conditioner variants (6 of 7 variants), about half of chewing gum variants (5 of 9 variants), and few toothpaste variants (3 of 9 variants). The colours expected by category buyers are presented alongside the colours prototypically used on pack in Table 1.
3. Do category buyer expectations match marketing practice?
To effectively operate as useful signals of product variety, the colours used by brands on-pack must match the colours expected by consumers.
Method:
To understand the degree to which these measures align, we calculate the proportion of prototypical colours on-pack that overlap with consumer expectations. For example, if charcoal toothpaste has two prototypical colours, black and brown, but category buyers only expect one colour, black, alignment would equal 50% as only one of the colours align. By comparison, if brands prototypically use light blue to signal sensitive varieties of fabric conditioners, but category buyers expect white, alignment would be 0% as none of the colours align.
Key results:
This research finds that the average alignment between brands’ use of colour on-pack and category buyer expectation of colour is just 16%. Alignment is strongest in the chewing gum category (26%), but still three in four are not aligned. In the fabric conditioner category only 14% of expectations correspond with the same colour on-pack, and for the toothpaste category this overlap is just 6%. A detailed summary of prototypical colours and consumer expectations is presented in Table 1.
Table 1: Alignment between Category Buyer Expectation and Prototypical Colours On-Pack
4. Is a picture really worth a thousand words?
Finding that colours perform quite poorly as variant signals, we next asked, do category buyers have consistent expectations of variant images? And, how do images perform, relative to colours, as signals of variant type?
Data:
To evaluate the comparative potential of images to signal variety on-pack, the same 1853 category buyers were surveyed across the three categories in the USA.
Method:
Respondents were presented with a single, greyscale image and provided with a matrix of greyscale, mock-up product images for different variant types for a fake brand. They were asked to select which, if any, variant(s) were best represented by the image shown. Royalty free images were used to avoid brand size or usage effects that could arise from using real pack images. For example, a picture of a palm tree with a coconut was used to represent tropical varieties.
Key results:
We found that images were linked to product varieties by more consumers for 23 of 25 variant types tested. In fact, 23% more category buyers held associations between images and variants than colours and variants (73% compared to 50%, on average). Each image was linked to only one product variety, indicating images are also more explicit signals of variety.
Although this research finds a clear, communication advantage of using imagery, in practice only 56% of products in our sample use images to signal variety. This result shows brands are creating abstract connections to flavour, scent, and functional characteristics with colour, opting to be creative rather than literal and sacrificing the ability of these on-pack design attributes to clearly signal the variety.
We acknowledge that the use of images may be less practical in certain categories, such as those with complex or highly specific variant types. Nevertheless, images demonstrate clear advantages for variant signalling due to their specificity. To leverage this advantage, brands must ensure variant images are prominent enough on pack to be easily noticed.
Summary and key take outs
It is common for product varieties to be prototypically represented by one or more colours on pack. Despite 80% of individual products using these colours on average, they do not appear to match the expectations of variant colour held by consumers. The colours used on pack align with the colours expected by consumers on only 16% of occasions. We find that:
- There is poor alignment between consumer expectations of variant colours, and the colours used by brands to signal variety on-pack.
This lack of alignment might be one of the contributing causes of new line extensions failures. To remedy the risk of a disconnect, brand managers should avoid inferring the packaging cues used by consumers based on common use of design attributes in their categories. Instead, these results highlight the need for more consumer-based category-specific research to develop an evidence-based approach to variant signalling.
There is better alignment for flavours or scents with a clear link to the natural environment. For example, purple for lavender variants, black for charcoal, brown for cinnamon, and orange and yellow for tropical fruits. Colours with a direct link to variants are likely to be more effective signals.
Images are more explicit signals of product variety than colours and may improve variant navigation in retail environments.
For brands launching a variant that is new to the category, the practical implication is to opt for imagery over colour to signal that variant’s properties where possible. This will reduce ambiguity for category buyers and prevent the fragmentation of colour-based Distinctive Assets that may be used to signal the brand on pack and elsewhere.
If launching a new range or a variant into a category with prevalent colour norms, brands should be aware that this prototypical use of colour on pack may not reflect the colours expected by category buyers. Therefore:
- The use of colour to signal variety is not without risk and can return varying degrees of success.
For brands already using a prototypical colour as a signal of variety on-pack, testing of consumer association is important. To improve within-range navigation, managers should educate consumers on the colour’s meaning. For example, by co-presenting the colour with an image to build the association between the colour and variant for category buyers over time.
Finally, it is important to remember that:
- Within-range navigation (i.e., signalling the variant) should not be at the expense of brand navigation (i.e., finding the portfolio among competitors on shelf).
An important question to pose when designing a new pack is, if a consumer were to see only this pack, could they easily link it back to the master brand? To assist with this, we recommend prototypical colours account for no more than 25% of the pack face, to reduce fragmentation of colour across the portfolio. One way a brand may use colour to signal variety, without jeopardising brand building, is to feature a variant banner in the same position on all packs, such that the design of the banner remains consistent across the portfolio, but the colour differs for each variant.
______________________________________________________________________________
Related Corporate Sponsor Report on Packaging Design:
Caruso, W., Romaniuk, J., Page, B., Anesbury, Z., Williams, J. (2023). Report 114: Lifting the Lid on Consumer Packaging Redesigns. The Ehrenberg-Bass Institute, University of South Australia.
______________________________________________________________________________
