Short-term effects do not predict long-term advertising success

Path into salt

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” is one of the most commonly quoted comments about advertising, being variously attributed to John Wanamaker and William Lever. Perhaps as a consequence, one of the key uses of market research is to test, monitor, and track advertising. However, it might well be that half of the money spent on testing and tracking advertising is also wasted.

How does advertising work?
In the distant past we used to think advertising worked along the lines of the AIDA model, it helped create Awareness/Attention, Interest, Desire, and Activation. However, more recent research, including behavioural science, econometrics, and media mix modelling, have shown that the picture is much more complex.

One of the best studies of how advertising works is one carried out for the IPA by Les Binet and Peter Field, which produced the report “The Long and the Short of It”.

Short Term and Long Term?
One of the key findings in the work by Binet and Field is that short-term success is not a good or reliable indicator of long-term success. Rational measures, such as standout and attention are quite good at predicting short-term effects (such as whether people will try something, click on it, etc). However, these measures are not good predictors of long-term success.

What is long-term success? Perhaps the best way of encapsulating long-term success is to say that it reduces price elasticity. If we become more attached to a product or service we keep buying it, even if the price goes up, i.e. we are less price elastic, i.e. it is directly related to the ability to make more net profit, not just to the ability to move the volume of sales.

If the short term is so bad at predicting the long term, why do we focus on measuring short-term predictors?
In my opinion, the key reason for focusing on short-term predictors is that we can measure them, and they tend to correlate well with short term results – and in today’s short-term world an immediate correlation is quite reassuring.

The problem with long-term measures is that there is no clearly established research technique that predicts the long-term effects of advertising, although many agencies are working hard to find solutions. There is a feeling that focusing on emotional messaging might be capable of being predictive of long-term results, but the jury is out at the moment.

Watch the video
You can hear Les Binet and Peter Field talk about their findings in the video below.