Do perceptions drive sales or sales drive perceptions for your brand?

Screen Shot 2018-08-15 at 8.35.07 AM

Based on both our experience and marketing theory, brand perceptions can drive brand purchase behavior, and brand purchase can drive brand perceptions. From the earliest ‘purchase funnel’ to the latest ‘online decision journey’, marketers proposed that consumers form a consideration set and gave the included brands preferential attention and buying likelihood. Other factors also influence purchase, including distribution coverage and prominence, and point-of-purchase actions such as price promotions and displays. Likewise, purchasing and using/consuming a brand affect consumer perceptions of it, leading also to word-of-mouth that changes other consumers’ perceptions.

What has been the subject of debate is the relative importance of brand perceptions driving sales vs sales driving brand perceptions. This is key to advising managers on brand resource allocation: the former calls for marketing to persuade prospective customers with targeted and creative communication, the latter for spending on distribution coverage and mass market awareness. On one extreme are ‘brand consultants’ who focus all their efforts on obtaining the best ‘positioning’ in consumers’ minds and trust this will translate into sales. On the other extreme is Byron Sharp, who claims in yesterday’s blog (https://byronsharp.wordpress.com/) that ‘we have practically no knowledge of how much particular perceptions affect behaviour’ and that ‘Sales (i.e. behaviour) strongly affects perceptions, so correlations between the two are largely, if not totally, due to behaviour causing the perception.’ Both extremes meet in the lack of scientific evidence for their claims, which our research aims to provide.

Scientific evidence of brand perceptions changing before consumer behavior (sales) change has been well documented in recent literature. Srinivasan, Vanhuele and Pauwels (Journal of Marketing Research, 2010) show this across fast moving consumer good (FMCG) industries and also demonstrate that including brand perceptions improves sales explanation even in a model that accounts for long-term effects of marketing actions (including advertising and distribution).  Pauwels and van Ewijk (Marketing Science Institute Best Paper award, http://www.msi.org/reports/do-online-behavior-tracking-or-attitude-survey-metrics-drive-brand-sales-an/) extend this analysis to 15 industries, including cars and service subscriptions and include online consumer behavior metrics (paid, owned and earned media) to show dual causality between them and survey-based attitude metrics. Scientific evidence of consumer behavior (sales) changing before brand perceptions change, has only recently been uncovered, mostly in unpublished studies. For one, my analysis of Spanish FMCG data over the 2000-2010 period showed that sales for diet Coke dropped almost immediately after Coke Zero was introduced. However, attitudes for diet Coke stayed up for several months before they followed the sales decline. Consistent with Byron Sharp’s view, I interpreted this as a consumer need for attitude-behavior consistency: ‘as long as there is still some diet Coke in my fridge and/or I still remember enjoying it, I will state I like it when asked in a survey/interview about my feelings’. However, after I forgot about diet Coke, I will state I don’t like it to be consistent with my behavior of not having bought it for months.

In a current analysis of 153 worldwide brands, we find 35 with brand perceptions changing at least a month before sales change, 30 for which sales change at least a month before brand perceptions change, and 11 brands with dual causality. The latter are personal care brands, so we are currently analyzing how brand and category factors influence these relationships, hoping to give managers specific guidelines. The theory relates to different levels of category involvement: for many grocery products, consumers mostly decide at the point of purchase, while for more involved purchases (such as personal care), a change in attitudes is necessary for changes in buying behavior (Alba, Hutchinson and Lynch, Handbook of Consumer Behavior 1991).  I don’t see how the truth is served by Byron Sharp’s unfounded and absolute claim that ‘This powerful causal relationship makes quantifying how particular perceptions drive other perceptions or sales impossible.’ Let’s make the impossible possible. For your brand.

Advertisements

7 thoughts on “Do perceptions drive sales or sales drive perceptions for your brand?

  1. Koen, you know I’ve published many articles on the importance of mental availability in determining market share. So it’s disingenuous to imply I was saying awareness and brand knowledge is worthless. My post was about brand image perceptions. Marketers track these perceptions and try to infer some meaning in small shifts in specific image attributes (e.g. we are up a bit this month on “cares for the community”, but down a bit on “good business partner”). One of my points is that we have almost no scientific (i.e. generalised, under known/documented conditions) knowledge to guide such brand image interpretation. None of the articles you cite provide us with such knowledge, or am I missing something?

    • thanks, Byron, you know I spent over a decade quantifying the effect of such brand image perceptions on sales, e.g. ‘What is important: Identifying Metrics that Matter’, JAR 2009 and Chapter 8 in my book ‘It’s not the Size of the Data, It’s How You Use It’ (AMA, 2013). So my key objection to your blog was the claim that it is “impossible” to quantify how particular perceptions drive sales. You are correct that the empirical answers to these questions often disagree with the manager’s opinion. A few examples: I advised Nissan to ‘only’ spend $100M ad budget on launching the Leaf in the US, not the $250M they wanted in an over-calculation of how improving the ‘green’ image perception was worth for their umbrella brand and other product sales. In an emerging country, I consulted a car manufacturer who built the 2nd largest Facebook following in the industry and spent 25% of his full budget on it. Sales did not budge however, and my analysis showed his following did not increase test drives, which was a strong and leading KPI of sales. In a final case, managers were upset because, while sales and profits were up, their particular brand image perception metric was down – only there was 0 evidence it was leading sales.

      TLDR: I believe managers and their companies are best served by analyzing which specific perception metrics are leading KPIs for their brand – so I fully agree with you in lamenting they typically don’t. This is unwise not just because much brain power and money is wasted, but also because each of my studies has found at least 1 perception metric that is a leading sales KPI and that can be affected by marketing – and thus allows managers to quantify the link and get more funding for their best communication ideas (see ‘Consumer Attitude Metrics for Guiding Marketing Mix Decisions, Marketing Science 2014, for these criteria and examples). Current projects involve industries as diverse as construction, drugs and deodorant, and I don’t expect to find exactly the same drivers.

      Did we obtain any generalizable knowledge across cases? (1) advertising awareness is often a driver, even controlling for eg aided and unaided brand awareness (JAR 2009) – I believe this supports the salience argument you make based on your research, (2) unaided awareness matters most in high involvement categories, aided awareness in low involvement categories, and (3) different perception metrics matter for high end versus low end brands: page 107 (book, 2013) lists ‘entertaining friends’ and ‘snacking on the go’ usage occasions as key drivers of consumers buying more national brand vs private label, consistent with conspicuous consumption. Looking forward to working with you and others to examine whether these findings hold up!

      • Yes we do agree on much. In spite of coming at things from a different perspective, which I think is the chief cause of the recent misunderstanding. I was writing about brand image perceptions (what the brand evokes) not mental availability type metrics which I’m on record of supporting.

        And I was writing about the sort of generalised knowledge that lets us make predictions, eg because we know “in this situation, this image perception matters this much”. My blog post pointed out that we don’t have hardly any of this sort of scientific knowledge. This is different from having to go do a multivariate analysis for a specific brand and market that suggests that there has been (in the past year or so) a causal affect of movements in this perception and sales. I’m also sceptical that such modelling yields reliable predictions… looking forward to seeing such tests.

        I’m interested in the generalised knowledge you list above.

  2. Pingback: Mark Ritson: Byron Sharp is wrong - brand perceptions influence sales

  3. Pingback: Why read ‘The Book of Why’? Pearl, Pipes & missed chances in Time | Smarter Marketing with Better Results

Leave your comment here

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s