Bringing precision and elaboration into ad testing

ximagination © – 123RF.com

Everybody knows that you can’t research ads in focus groups, and no researcher of any standing will openly admit that they do. But that doesn’t stop brand managers and account executives from talking about the take-out, insights or findings that stemmed from a couple of focus groups.

This is not a quallie-bashing article – ‘caveat emptor’, after all – clients have the right to do silly things.

What this is about is sharing some of the work that we’ve been doing for the past 25 years and to argue for a better way for it to be used collaboratively and constructively to understand what drives the business value of advertising, without criticising the creative execution.

There are four important components and these will be discussed in four ‘episodes’ here on Bizcommunity.

The four sections will be:

• The objectives of advertising testing
• The measurement of advertising
• The consumers of advertising
• How it all fits together and case studies

These thoughts were presented at the Amra conference in Johannesburg in February 2017.

Research

Who is winning in Africa, and is Africa winning?

African Market Research Association  8 Feb 2017

The objectives of advertising and implications for ad testing

The point of advertising is simply to implement a business strategy – so it would seem to make sense to factor that into the equations at some stage when evaluating advertising or individual advertisements.

Kantar Millward Brown recently announced their annual love fest of ads that are the most liked. So, why would a marketer be pleased that their ads are liked? Well, it is founded on work published in the US in the early 1980s, principally Mitchell and Olsen (1981) and Shimp (1981), where they write about the pivotal role of Attitude to the ad – generally referred to as A_Ad. It’s defined as a predisposition to act on the exposure of an advertising impression. As such, it’s easy to slip into the presumption that liking an ad means the ad is likely to be effective.

Advertising

Kantar Millward Brown announces South Africa's Top 20 Best Liked Ads of 2016

Kantar  8 Jun 2017

The fact is that examples abound where this turns out to be untrue. One of the best examples was the babies in executive chairs in the Sanlam campaign in the 1990s. Impact Information – which became Millward Brown – reported that it was the most-liked advertising campaign for decades. It also led to the responsible agency losing the business to Berry, Bush, di Bella & de Villiers within two years. So what went wrong? I happened to be doing ad testing work with Berry Bush at the time and this was one of the campaigns that we researched. Our conclusion was that:

• a) The advertising was truly effective – as advertising, but,
• b) As an implementation of business strategy, it was fatally flawed.

In researching the field in depth, we found that academic articles about A_Ad proliferated in the years after 1981, seemingly peaking in the early 90s when things became tangled up with psychology and physiology, heading towards neuroscience.

Personally, I felt that the academics were getting further and further away from the practical matters of advertising development and optimisation. I did, however, find one paper that started discussing the antecedents of A_Ad. The most useful paper was by Lutz, RJ; MacKenzie, SB; & Belch. GE (1983), titled: ‘Attitude toward the ad as a mediator of advertising effectiveness: Determinants and consequences’. Advances in consumer research. In plain language, this paper is about what A_Ad actually means. Its conclusion was that there are six basic dimensions to A_Ad. The importance of this is that these six factors are the key to understanding how to measure A_Ad. In the same way that attitude to the brand (AKA brand image) is determined by measuring the importance of the attributes associated with the brand, so it would be possible to measure ‘liking’ (the translation into English of A_Ad), by measuring the importance of these six dimensions.

The following are my translations of the dense academic language that is required for acceptance for journal articles. A different reader may come to different conclusions and I’d be very happy to converse with anybody who has studied the article and can debate the interpretation have settled on. The six dimensions that predict ‘liking of advertising’ are:

1. Attitude to the brand: The fact that people like advertising for the brand that they are entrenched with has been shown in a number of papers generated from the literature surrounding the conversion model developed by Butch Rice and Jannie Hofmeyr.
2. Attitude to the company: Many companies have a dominant company brand, among them most pharmaceutical, technology, financial institutions and some FMCG companies (Nestlé). The same principle as above applies, in that the advertising of a company would be liked by those who are favourably disposed to the company that is advertising.
3. Attitude to advertising: If a person does not like advertising in the category, then they are less likely to pay attention and advertising will have less effect on them. And of course, the converse applies which implies that a category could have an advertising genre.
4. Relevant and useful information: In most categories, buyers are looking for answers to questions. This doesn’t necessarily only apply to technical products but also role and behavioural cues from cosmetics to alcoholic beverages.
5. Improve your mood: There’s no question that a great deal of advertising is entertainment, even if only to attract the attention of buyers. The extent to which an advertisement attracts, entertains, helps one feel better is required more in some categories and not in others.
6. Truthful and believable: This relates to the extent to which the communications are reliable to be acted upon. In financial services, for example, hyperbole and irresponsibility are likely to have less effect. Referring back to the Sanlam babies’ campaign, this was the only dimension of these six that the campaign scored very low on.

Evaluating advertisements from the perspective of these dimensions should help gain a more textured understanding of what drives advertising effectiveness. Along with this understanding is the need to evaluate the effectiveness, which is why we have developed an automated product called Ad-Audit.

For more information, please contact me at PSA on 083 255 2668 or click here to send an email.

The most important step is how one can measure these and if there is an optimal blend. This will be discussed in the next article.