Testing How Brands Grow

In my 2 previous blogs, I summarized the top 10 actionable insights from ‘How Brands Grow Part 2 Revised Ebook’ (HBG) for established and new brands. But if your main thought was ‘How does this differ from advice other smart people give in marketing?, this blog’s for you. I focus on four points that have seen considerable debate, and offer 4 Research Questions (RQ):

1. Consistent attribute positioning versus changing focus on Category Entry Points?

David Aaker recalls pushing back on Burger King’s insistence to become known for breakfast, leading the CEO outside to point at the sign and say: ‘what is your company’s name’? ‘Burger King’, ‘then shouldn’t you instead aim to become known for the best burgers?’. Throughout marketing literature, authors have advocated for a consistent positioning over time on one or a few main attributes (Braun-LaTour and LaTour 2004, Erdem and Swait 1998, Ries and Ries 2002). Instead, HBG calls for changing the ‘what your brand is known for’ over time to add category entry points, consistent with what Burger King was aiming to achieve in the early 2000s. So was Burger King successful in doing so and/or should it have devoted its resources to communicate its lunch/dinner burger? At a more general level, we found evidence that younger brands should be consistent over time, but older brands should switch positioning towards attributes that matter more to consumers at the time. However, we did not look specifically at category entry points.

RQ1: Which brand grow fastest: the one with a consistent positioning over time (e.g. on 1 Category Entry Point), or the one that changes focus to add Category Entry Points?

2. Advertising pulsing or always on?

The advertising budget can be spent evenly over time or in ‘pulses’ of a certain frequency, length and predictability. The review by Vakratsas and Naik (2007) shows that early studies consistently find that an even advertising spend is better, consistent with Sasieni’s (1971) result that no pulsing schedule can outperform the even schedule. However, copy wearout and repetition wearout imply that pulsing is best. Interestingly, following HBG’s advice implies both wearouts are unlikely to apply:

• Advertising mostly moves light users, which are unlikely to pay attention to the ad in the first place, let alone get annoyed by it;
• The budget should maximize reach, so a frequency of 1 means no repetition wearout.

Recently, Gijsenberg and Nijs (2019) show that investing in top-of-mind awareness through maintenance advertising insulates brands from competitors’ actions and boosts sales in a variety of real-world scenarios. In their study, no brand engages in even spending, but many in flighting (57% switch between zero and one-week high advertising) or pulsing with maintenance spending (39% switch between high and low advertising), as illustrated in their below figure.

Soft-drink and cleanser brands A pulse with maintenance, while soft-drink and cleanser brands B and C pulse without maintenance. The former perform better, with spending about 1/3 of the budget on maintenance getting the highest sales lift. However, performance differences among pulsing-with-maintenance scenarios and with the even spending scenario were not statistically significant. Moreover, the authors analyze only consumer packaged goods.

RQ2: Under which conditions is even marketing spending better than pulse-with-maintenance and how large should maintenance be as a % of the total budget?

3. Only reach or also frequency of ad exposure?

On how to spend your advertising budget, HBG is clear: ‘as much frequency as your budget can buy’. The oft-cited ‘3 times frequency is optimal’ is called a myth. This is consistent with generalizations about advertising effectiveness (Tellis JAR 2009): ‘increasing the frequency of ad exposure increases brand choice at a decreasing rate’ and add ‘established brand have an earlier peak response than new brands’. I also find that the second exposure has a lower impact than the first for low-involvement products, and that ‘3 times’ is a myth, but for other reasons. The latest research (Burton et al 2019) finds that even ’10 times frequency is too much’ is a myth, as ‘consumers who saw an advertisement 10 or more times had greater purchase intentions than consumers with less exposure’. This is consistent with my research that the highest purchase happens for consumers exposed 15-20 times to online ads. So there is conclusive evidence for diminishing returns to frequency, but not for supersaturation (consumers lowering their purchases because of excessive ad exposure).

Hence the practical question: is it best for your company to spend the next dollar on reaching the same customer again or to reach a new customer? It all depends on diminishing returns in reach: how much benefit will you from exposing an additional consumer? According to HBG, there are no diminishing returns to reach: as long as you are touching a category consumer, the behavioral impact is similar. That’s not my experience in consulting companies: they get less behavioral impact per exposure once they go beyond their identified target segment. So let’s test this. My hunch it depends on category involvement. For a new chewing gum, 1 exposure suffices and reach is king, For a device, such as the Echo, consumers have to be touched several times before purchase, and some consumers are more likely to respond than others.

RQ3: How strong are diminishing returns in advertising reach and frequency, and which category and brand characteristics does it depend on?

4. Distinctive assets or attitudes and differentiation

While HBG rejects building consumer attitudes in general, and brand differentiation in particular, as an achievable and desirable goal, the book promotes building distinctive assets (such as characters, colors and archetypes) that link your brand to category entry points and thus give it mental availability. To many, the distinction may seem like hair-splitting, but it does have several strategic consequences:

(a) brand-building strategy: Brand-building efforts and investments will be very different whether an organization aims to foster specific differentiated attitudes regarding its brand(s) or whether the focus is on building salience to induce transactions.

(b) marketing research: little diagnostic value in mindset metrics, beyond a reflection of sales figures, means little practical value of large parts of the market research industry and changes the skills sought by organizations and taught at universities.

In our forthcoming ‘Why Brands Grow’ paper, Oliver Koll and I find that consumers do report meaningful differences between brands, and that this differentiation is the main driver (among the studied variables) of penetration and of satisfaction, and a key driver of market share:

Interestingly, the differentiation impact is larger for emerging markets (we studied Saudi-Arabia, Thailand and Indonesia versus UK and Germany) and for smaller brands:

While this research studies 153 brands in 5 countries, it has several limitations, including that we don’t have data on HBG metrics such as mental availability (as opposed to brand awareness), distinctive assets and category entry points. Moreover, we can’t control for many other factors on which brands may differ, enabling a propensity matching analysis. So the question for future research becomes:

RQ4: Is it better to build differentiation or to build distinctive assets for your brand and how does it depend on country, category and brand characteristics?

Do you have answers or data that can help us answer these research questions? Please let me know!