I've been reading a very excellent issue of the Journal of Advertising Research that the lovely Gwen at the ARF sent me.
It's a special issue that concerns itself with a very important question [for those of us in advertising]:
What do we KNOW about advertising?
If this seems to be an obvious question, ask yourself to define clearly what advertising is, and exactly how it works, in all cases.
It's tricky this business of ours. That's what makes it so interesting.
[PS> If you don't have some kind of model of how you think advertising and brands and that works - go and think about if for a while , because this is important.
If you don't have some idea HOW and WHY things WORK, how can you hope to make things that do?
If you don't have some model in your head about how putting pieces of content into media [without even getting into the more complex world we live in now] can create business value for companies, by which mechanism this process happens, commercially, culturally and psychologically...well you should.
And then you need to recognize that no one model will hold true for everything:
No single theory or
group of theories can explain it all, because advertisements work in such
different ways. There is no point in looking for an overall theory.
So maybe go get this issue of the JAR and start there.]
[That was quite an extended parenthesis wasn't it? This one is much shorter.]
One of the research papers I really enjoyed was from measurement masters Les Binet [DDB Matrix] and Peter Fields [my measurement mentor].
It highlights some of the findings from a meta-analysis they undertook for the IPA in the UK, using the IPA DataBANK, which holds all the winning papers for the IPA Effectiveness Awards [one of the few awards which focuses exclusively on the business results of advertising].
The IPA Effectiveness Awards are recognised by agencies and clients as Adlands most rigorous awards scheme because entrants have to prove to a jury of experienced clients that their communications strategies have worked in hard business terms. These awards are open worldwide to agencies, media owners and advertisers.
The analysis looked at 880 winning papers and tried to draw some general conclusions about what seems to correlate strongly with business success.
The paper is called:
Empirical Generalizations about Advertising Campaign Success
You really should read it.
A couple of points they highlight:
"Campaigns that aim to reduce price sensitivity are more effective than campaigns that aim to increase sales or market share."
"Campaigns that focus on customer acquisition are far more effective than campaigns that focus on loyalty."
[Whilst they don't mention this there is an obvious reason for this - loyalty campaigns are forced to compete with direct experience of the customer, whereas acquisition campaigns are not.]
"Campaigns that have the stated aim of getting the brand and the marketing talked about are particularly effective."
But probably my favorite bit is the bit I put up top. With the exception of direct response advertising, which seems to work better with rational messages:
"The most effective advertisements of all are those with little or no rational content." Les Binet/Peter Field
I especially love this because it validates something I posited in a previous post:
"This effect, where commercial grammar overrules social grammar when both are employed is detailed in Predictably Irrational - it further supports Feldwick's view below in the sense that specific product messages would frame communication as commercial and therefore negate the development of social relationship building, or brand preference.
This is probably contentious, but it suggests that if you attempt to employ any rational messaging in brand building communication, you might well be shooting your brand in the foot". From Social Gravity and the Value of Free new emphasis]
So this piece of research, if we hold it to be valid and that, alongside Feldwick's paper, requires those who still hold on to things like Reasons to Believe and USPs to have a good hard think.
I'm not saying this is the answer or anything, but it does really need thinking about.
As Mr Feldwick has pointed out, most advertising professionals will intuitively agree, but in practice we find it really hard to accept that this is how things work, and so default to doing the opposite.
Kindly, the ARF have granted me permission to the post the research paper here.
Download BINET & Field IPA DATABANK Empirical Generalizations about Advertising Campaign Success.