How DOES advertising work then?

Posted by on Feb 8, 2012 in behaviour change, Market Research | 8 Comments

For adplanners and the like:

In the pre Christmas (holidays to you, Gareth) rush, you might have missed this major U-turn by the lovely folks at Millward Brown research (the people who brought you the "awareness index" a key metric that assumes that advertising achieves its effect primarily by impacting memory – awareness and recall). 

Suddenly – after years of dissing those like my chums at Brainjuicer – who dared to suggest that emotions were the key to advertising effectiveness, the folk we used to know as The Leamington Spa Gang have turned 180 degrees. Apparently, emotion is everything or pretty much everything.

Nice to have you aboard, chaps. Will you be dumping AI scores then?

 

8 Comments

  1. Gareth
    February 9, 2012

    Oh my lord, I hadn’t seen this. They buried it in the holiday break. It’s like the Pope declaring there is no heaven. Wow.

  2. Graham Page
    February 10, 2012

    I think you’re misunderstanding Millward Brown’s position on both emotion and advertising memorability. Memory is of course still important in marketing effectiveness as most influence doesn’t happen at the point of exposure to campaigns, but later on, as brand equity interacts with point of sale factors and product experience. Brands are nothing if not a function of memory. Emotional response to marketing is important for two reasons – it affects whether people like the brand, which frames all their decision making about it and is a powerful shortcut to purchase; it also is a really powerful driver of memory as we are wired to remember things that make us feel strong emotions.
    A fairer criticism is that Millward Brown haven’t always called out the importance of emotion explicitly enough. But measures of emotion have always been part of the metrics. One of the biggest drivers of the AI is and always has been people’s emotional response to the campaign – which has been integral to the Link methodology since 1989. One of the reasons why Polly and Dom were able to write the paper you refer to was that emotional metrics have been collected for many years. However, MB has been talking directly about importance of emotion for a long time now – for instance my paper dating back to the ESOMAR congress in 2005. And the inclusion of implicit measures of emotional response in MB’s research offer – such as reaction time measures and automated coding of facial expressions – shows there is a real commitment to measuring emotions, and using the findings to help clients create better campaigns and brands.

  3. Mark Earls
    February 10, 2012

    Thanks for response, Graham but I’m not sure you’re on as firm a foundation as you think. Seems to me what you’re doing is just pushing two models together here (and not in the way that for example Hall and Maclay’s original framework did): rather you’re stretching a model which is based on high-level cognitive variables (esp. memory) to try to take in the new ground of emotion that is associated with much lower level cognitive variables.
    The AI has to jump one way or the other, surely? Either it’s about memory and awareness or it’s not. You can’t have it both ways, really?

  4. Graham Page
    February 10, 2012

    Thanks for the feedback, I don’t see the conflict. Events which are emotionally powerful are memorable. And a high AI doesn’t mean that you have actively ‘thought’ about an ad’s message. In fact, the AI and very cognitive measures along the lines of ‘does this ad make you more likely to buy the product?’ are not related. Many people seem to think that the AI is about explicit message recall, but it isn’t and never has been.

  5. Mark Earls
    February 10, 2012

    Thanks, Graham.
    It is worth restating – for anyone not acquainted with your tools -that the AI and propensity to purchase questions are not connected
    I’m still struck by the fact that you see memory (whether its recall or recognition) having a role to play at all. Why would that be? It seems a very unreliable and high level cognitive function (as typically explored in survey research.
    Perhaps you could explain why?

  6. MR Avenger
    February 13, 2012

    In Admap July/August 2006, IPSOS ASI published a paper, using their ad database to explore this area.
    Their analysis shows that ‘Ads processed with ‘ultra’ low attention…have a significant impact on brand choice.” (They use the term ‘attention’ but they are talking about memory: ultra low attention was ‘those who neither recall nor recognise the ad, but whom we know have been exposed.’)
    What is really interesting, though, is their conclusion that “Ads that succeed in being processed with high attention are more than 2.5 times more impactful than ads processed with low attention, and six times more impactful than ads processed with ultra low attention.”
    The IPSOS ASI chaps go on to say that their research…”suggests that LAP is a slightly more potent force for information-driven ads than for the less informative, more fun and emotionally driven ads.”

  7. MR Avenger
    February 14, 2012

    On the role of memory:
    In Admap July/August 2006, IPSOS ASI published a paper showing that ‘Ads processed with ‘ultra’ low attention…have a significant impact on brand choice.” (They use the term ‘attention’, but they mean ‘memory’; ultra low attention is ‘those who neither recall nor recognise the ad, but whom we know have been exposed.’)
    What is really interesting, though, is their conclusion that “Ads that succeed in being processed with high attention are more than 2.5 times more impactful than ads processed with low attention, and six times more impactful than ads processed with ultra low attention.”
    The IPSOS ASI chaps go on to say that their research…”suggests that Low Attention Processing is a slightly more potent force for information-driven ads than for the less informative, more fun and emotionally driven ads.”
    In November 2007, Millward Brown published an article in Admap which showed “a clear relationship between the AI prediction and the proportion of cases seeing a sales share rise.”
    You really should read Admap!

  8. Leigh Caldwell
    March 24, 2012

    I’m coming to this conversation rather late, but memory is a broad capability which covers various phenomena. One example is conscious recall of facts or rationales (as in survey research), but unconscious familiarity with a brand or product, and pattern recognition, are also memory effects. After all where does the familiarity reside if not in our memory?
    I think we are all probably in rough agreement that those low-level, less conscious effects are more important. They are indeed strongly influenced by emotion. I don’t know how accurately AI measures that level but it sounds like that’s what Graham is talking about.