Advertising reach and frequency

In this instalment of ‘Best Practice’ Colin McDonald examines the issue of reach (or cover) and frequency of advertising exposure.

Advertising Reach And Frequency

Colin McDonald

'Reach' (Cover in US) and 'frequency' are easy to define. Reach means the proportion of the target population who will see our campaign at all, 'frequency' means the number of times those people will see it. Advertising schedules aim to). The term 'effective frequency' denotes the 'right' level of frequency, and effective frequency planning (EFP) has been the basis for most schedule modelling for the past 30-odd years. Unfortunately, this simple account begs some key questions (, ). First, frequency of what? Are we talking about opportunities to see/hear (OTS or OTH), ie exposure to the medium containing the ad (usually the only thing for which there is any available measurement) or proved exposure to the ad itself? We know that proved exposure is a lot lower than OTS, but how much lower may vary between different media and products. Secondly, frequency when? Two exposures may have very different effects if they are concentrated within a day, or close to the next shopping occasion, than if they are spread over weeks. Thirdly, what effects (sales, purchase probabilities, awareness measures and such like) are we to use to compare frequencies - they may all be different? Underlying these questions is continued uncertainty - and disagreement - about the psychology of how people respond to ads in different circumstances (). Much of the argument in recent years has arisen from people trying to fit limited evidence into different all embracing theories. Rise and fall of the 'magic number' (, , , , ) Before EFP, planners worked with average frequencies, which were flawed by the fact that many different media exposure patterns could lead to the same average figures. When computers appeared in the 1960s, it became possible to work with frequency distributions. Planners could now see how many people would be reached once, twice, three times, and so on. But they needed guidance to use this knowledge: what was the 'right' level? A rule of thumb was very attractive. If 3+ (say) could be agreed as the 'effective' frequency, then scheduling plans could be compared on the basis of which one delivered the best 'effective reach' (ie maximised reach at 3+). Effective frequencies of two or three are intuitively appealing, since they assume a threshold that a new ad must break through to get noticed (, ). The idea was given weight by Krugman's 'Three Hit Theory' (, ): the first exposure simply gains attention ('what is it?'), the second establishes relevance ('so what?'), and the third consolidates, either positively ('for me') or negatively ('forget it'). Subsequent exposures simply reinforce the third, unless there is a long interval, in which case one goes back to the first level. Empirical support was found in McDonald's study of switching behaviour (), and other evidence gathered together in Naples' 1979 book (), which seemed to show that frequencies of 2-3+ were indeed 'effective' in influencing actual purchases. Critics attacked both the theory and practicality of this rule of thumb. The research evidence showed that thresholds are rare. The most common response curve is 'concave downwards', or diminishing returns (, , ) - the first exposure has the greatest effect. Krugman's finding may be relevant in learning situations, but most advertising seen by potential buyers is for something already very familiar. Jones's analysis of single-source data () appeared to confirm strongly that diminishing returns are normal, and gave immediate support to 'recency planning'. Recency planning is often presented as an alternative to EFP, but is really a special case of it - the objective is to maximise reach, even with a frequency of one. The recency argument is that ads work when people are in the market to buy, and there are always some people in the market, although one never knows who or when. The objective, therefore, is to be present as continuously and widely as possible (, , ). With any plan (even for 1+ cover) a frequency distribution is inevitable: if you plan for higher frequencies, you cannot afford more than occasional bursts and the highest frequencies go to the heaviest media users (who may well be your heavy buyers anyway), thus wasting the extra money, 'a schedule bought to reach many viewers twice will waste most of its impressions reaching heavier viewers four, five and six times' (). Moreover, too many advertising impressions can be counterproductive, for creative reasons ().