Market tests: sales effectiveness of TV + magazines
A growing number of market tests and econometric analyses are proving that mixed-media campaigns involving magazines can sell products - and sell them more effectively than a campaign using television on its own. Some of the following examples are from the UK, while others are drawn from elsewhere and show that the results arise from the nature of the two media and not from any peculiarity of the UK market.
“Sales Uncovered”
In recent years PPA has commissioned two major effectiveness studies utilising the TNS Superpanel, “Sales Uncovered” and “Proof of Performance”. Each included, among the brands examined, campaigns which combined television and magazine advertising.
The “Sales Uncovered” study of 2005 [71], described in an earlier section, analysed 20 fmcg brands. Of these, seven were TV+magazines campaigns whose impact was assessed by medium. The average budget split across these brands is shown in the pie chart. 70% of the budget was spent in television and 22% in magazines.
For each of these seven campaigns, Superpanel main shoppers were ranked according to their weight of exposure to the magazine advertising and, separately, to the TV advertising:
- For the magazine campaigns, the top 40% were defined as the heavy or ‘exposed’ group (as a shorthand); in general, they accounted for about 90% of total magazine exposures. The bottom 40% in the ranking were defined as the ‘non-exposed’ group and acted as a control group; they accounted for around 2% of total magazine exposures.
- For television, a medium with a more diffused audience, the 40% most heavily exposed to the television advertising were defined as ‘heavy viewers’; they accounted for about 72% of all television exposures. The bottom 40% were defined as ‘light/non viewers’; they accounted for about 13% of all television exposures.
Aggregating across the seven campaigns, main shoppers who had seen none or very little of either the television or magazine advertising showed only a small increase in sales during the campaign period: 3.9%.
By contrast, those exposed to the magazine advertising but who were only lightly or not exposed to TV, showed a dramatically higher increase in sales. The same was true of those heavily exposed to TV but not exposed to magazines. For the two groups, the sales increase was 26%-29%. Clearly, advertising is effective in increasing sales.
Examination of the sales value figures showed that these two groups had similar purchasing levels during the pre-campaign period, and identical absolute increases in sales during the campaign period (hence the very similar percentage increases in sales).
It is notable, then, that magazines accounted for a much lower proportion of advertising expenditure than television: 22% of the budget, compared with 70% for television.
The relative cost-efficiency of the two media may be examined across all four groups in the bar chart above. To do this, the absolute increases in sales value (£) are profiled across the four groups as represented in the second pie chart.
Television advertising was closely linked to 71% of the sales increase: that is, the 46% among the TV+magazines exposure group, plus the 25% from the TV-only group. There will also have been some smaller effect among the group heavily exposed to magazine advertising and only lightly or not at all to TV.
Magazine advertising was also closely linked to 71% of the sales increase: the 46% plus the 25% from the heavy magazines & light/non TV exposure group.
This can be compared with the profile of advertising expenditure (first pie chart). While magazine advertising appears to have achieved something approaching the effect of television advertising, it did so at less than a third of the cost.
This does not mean that magazines are two or three times more cost-effective than television in all circumstances. What it indicates is that, pound for pound, magazines are much more cost-effective at the relative levels of expenditure in these seven campaigns. The reason is surely that television has been allocated too much of the budget and magazines too little. We know that diminishing marginal returns sets in for all media, and the expenditure on television in these cases appears to have gone past the point of severe diminishing returns. If, however, a more equally balanced amount had been allocated to magazines and television I would expect the two media to become much closer in cost-efficiency. I draw the conclusion that 22% is too low a share of budget for magazines.
The fourth column in the earlier bar chart titled ‘Magazines and TV: % increase in sales (£)’ shows the sales increase among Superpanel main shoppers who were exposed to both television and magazines. The figure of 30.3% is higher than for either the TV-only or the magazine-only bars on the chart – as one would expect, because of the advantages of mixed-media scheduling discussed previously. Indeed, one might have expected the difference to be greater.
However, granted the conclusion that the television advertising has passed the point of severe diminishing marginal returns, and that the magazine advertising is far from having reached that point, it seems likely that a more evenly balanced expenditure profile would have boosted the sales increase among this TV+magazines group of main shoppers, raising the height of that fourth column.
This topic is discussed further in section 41.
“Proof of Performance” : TV+magazines
PPA’s "Proof of Performance" analysis of the TNS Superpanel [74, 75]examined the link between mixed-media advertising and short term gains in brand share.
| 10 magazine+TV brands | |||||
| Brand shares (indexed) | |||||
| Magazine | All | Months* with magazine ad spend of: | |||
| category | Months | None | £1-50k | £50-125k | £125+k |
| Heavy readers | 100 | 98 | 99 | 101 | 109 |
| Light readers | 100 | 102 | 94 | 95 | 106 |
| Non-readers | 100 | 100 | 96 | 97 | 104 |
| Total panel | 100 | 100 | 97 | 98 | 107 |
*Magazines time-lagged by one month, to allow approximately for build-up of magazine reading. This analysis was run before NRS readership accumulation data were available.
For each of the ten brands covered, the ad expenditure in television outweighed the magazine expenditure, and on average the TV spend was about twice the magazine spend. The results are summarised in the table above.
The influence of the television advertising was felt by all the magazine exposure groups, for the brand share indices for all groups were higher in the months when magazine advertising was heaviest, which tended to be months when television advertising was also running. Nevertheless the effect of the magazine advertising can be seen in the gain of 11% among heavy readers (109/98) compared with 4% among non-readers (104/100).
This interaction effect can also be seen when the analysis is confined to those months in which television advertising was taking place. In the following table the ‘All months’ column is replaced by a ‘TV ad months’ column, and the indices in this column are slightly above 100, balanced by indices (not shown) of slightly below 100 for the months when no TV advertising occurred.
The much greater weight of television advertising during these months had the effect of reducing the variation between most of the cells in the table, but the main exception was the index for the heavy reader group in the months when magazine advertising was at its strongest. Here the brand share index rose to 114, a gain of 11% compared with the months with no magazine advertising (114/103).
PPA’s analysis neatly supports the Media Multiplier proposition that television plus magazines makes an advertising budget work harder than does television on its own.
| 10 magazine+TV brands | |||||
| Brand shares (indexed) | |||||
| Months when TV advertising was taking places | |||||
| Magazine | TV ad | Months* with magazine ad spend of: | |||
| category | Months | None | £1-50k | £50-125k | £125+k |
| Heavy readers | 104 | 103 | 103 | 100 | 114 |
| Light readers | 104 | 106 | 101 | 98 | 107 |
| Non-readers | 102 | 104 | 100 | 96 | 100 |
| Total panel | 103 | 104 | 102 | 98 | 108 |
*Magazines time-lagged by one month, to allow approximately for build-up of magazine reading. This analysis was run before NRS readership accumulation data were available.
USA: “Measuring Magazine Effectiveness” (MMA/MPA)
Magazine Publishers of America (MPA) commissioned from Media Marketing Assessment (MMA) an econometric analysis of MMA’s extensive seven-year database of marketing and sales information on 186 brands in 13 product categories, covering the period 1994-2000. The results were published in 2001 in the report “Measuring Magazine Effectiveness: Quantifying the Sales Impact of Magazine Advertising” [109, 110].
Central to MMA’s analysis was a measure of ‘effectiveness’, conceived as the sales effect each dollar has. Brand by brand, a ‘base’ volume of sales was modelled (sales that would have been achieved that year without additional marketing effort). The remaining sales above ‘base’ level were generated as a result of that year’s marketing efforts.
Each medium’s percentage contribution to these incremental sales was modelled, and divided by the medium’s percentage of marketing expenditure. This produced an effectiveness index. For example, if a medium contributed 30% of incremental sales and accounted for 30% of marketing expenditure, it would have an effectiveness of 1.0. The higher the index the better.
MMA found that magazines were substantially more cost-effective than either television or radio. Magazines’ effectiveness index of 1.2 contrasts with only 0.8 for television and 0.7 for radio. Expressed another way, a dollar spent in magazines produced on average 50% more sales than a dollar spent on television.
In another analysis, MMA examined how quickly magazine and television advertising individually generate incremental sales. Studying the weekly cumulative sales impact of each medium, MMA found that the two media work at very similar speeds. It was not the case, as sometimes suggested, that in general television works more quickly than magazines in creating sales.
UK: Cussons Carex Hand Wash
Before autumn 1996 Cussons Carex liquid soap had confined its advertising to television, with some support on radio and posters. But in autumn 1996 Cussons decided to test the use of magazines as part of a mixed-media campaign [120]. 81% of the budget remained on TV while magazines accounted for 19%, using TV weeklies, women's weeklies and women's monthlies.
Sales were tracked week by week using the TNS Superpanel of households. In the 12 weeks before the magazine campaign began, the Carex market share of sales was similar among households heavily exposed to the selected magazines and those lightly or not exposed. However as soon as the magazine advertising commenced Carex's brand share leapt among the heavily exposed households while being little affected among the light or non exposed households. This was maintained throughout the campaign period.
Additional research established that:
- Sales attributable to magazines were achieved at one-third of the cost of sales attributed to TV.
- Although magazines were only 19% of the budget, they added 50% volume sales above the uplift generated by television.
- The combination of magazines and television was found to produce a better return on investment than TV-only.
UK: Nielsen’s ‘Strategies of Successful Brands’
IPC Magazines co-sponsored one of the largest-ever studies into the long-term effectiveness of marketing activity. It was conducted in 1995-96 by Nielsen [106] and examined 300 products from 50 product fields, using the Nielsen Homescan consumer panel, Nielsen tracking data, and Register-MEAL advertising expenditure figures. For each product the market share, consumer penetration and loyalty were recorded for the six months ending April 1992 and the six months ending April 1995. Changes in these key brand measures were assessed against their advertising policy, pricing, promotions and innovation. Nielsen concluded that innovation is the best single means of developing the strength of a brand, and that sales promotion activity does not achieve brand building at all in the long term.
The lesson for media strategy was that, although many of these brands used only television for their advertising, on average advertisers obtained a higher brand share, and were more successful in maintaining or increasing share over the three-year period, if they used two forms of media such as television combined with magazines. Moreover brands using magazine advertising were on average both bigger and more likely to be growing. Nielsen concluded that “since magazine advertising is less expensive than TV advertising, this implies that magazines can be a highly cost-effective way of communicating with the end buyer”. In addition, the fastest-growing brands tended to be those with a higher proportion of their total adspend in magazines.
UK: Kenco Freeze Dried Instant Coffee
Kraft Jacobs Suchard, with a tradition of using television as a branding medium, ran a test of a television and magazines mixed-media campaign for their Kenco Freeze Dried Instant Coffee [121]. IPC Magazines was able to offer regional facilities in its publications on a sufficient scale, and the TV and magazines campaign ran from April to November 1995 in the London/South/Anglia regions which accounted for 35% of the market. In the rest of the country a TV-only campaign was run, on an equal expenditure basis. The budget for the magazine expenditure in the test area was found by switching a share of television money into magazines.
Among the target audience of ABC1 housewives, the net coverage achieved in the TV-only regions was near saturation but nevertheless it was improved slightly in the mixed-media regions. The mixed-media campaign also increased gross opportunities to see by 39%, improved the average frequency of exposure by 35%, and greatly reduced the cost per thousand exposures. Millward Brown tracking research showed gains in advertising awareness in the test area compared with the TV-only regions as soon as the magazine advertising began and it continued throughout the campaign period. Ad recognition levels were highest among magazine readers, and overall the reduction in TV spend in the test areas did not prove at all detrimental to brand image.
Most significantly, sales were improved by the mixed-media strategy. This was measured by two panels, Nielsen and TNS Superpanel.
With sales historically stronger in the south, it was important to allow for this in the analysis. Nielsen’s figures showed that prior to the test period Kenco’s share of instant coffee sales in the test region was 19.8% ahead of its brand share in the rest of the country. As a result of the test, this differential grew to 25.4%, a very significant gain of 5.6 share points.
The Superpanel was able to compare panel members exposed to the TV-only campaign with those exposed to the TV + magazines campaign. Results showed that the mixed-media campaign improved Kenco volume share by 7%.
Kraft Jacobs Suchard’s Director of Coffee Marketing, Nick Shepherd, stated “After careful analysis, we declared ourselves reassured about the potential for mixed-media advertising. Following the regional test results, we are using magazines nationally for Kenco this year – the surest sign that we believe it worked.”
Germany: Bauer and Hassloch BehaviourScan panel
In Germany a single-source panel has yielded further evidence about the virtues of mixed-media advertising. Bauer Publishing have been responsible for a number of tests using the GfK Hassloch BehaviourScan panel [122]. The panel consists of 3,000 households in the town of Hassloch whose purchases in a range of product fields are recorded using scanner technology in local stores. Panel members receive television through a cable system, which means that the commercials shown to each household can be controlled. Panel members also receive two weekly magazines as an incentive, and the advertisements carried in these can be varied too. For the launch of a personal care product two media strategies were tested, representing equal expenditures: some households received 100% television and others received advertising split 68% television and 32% magazines. After the campaign had run for one year the mixed-media strategy had outsold the TV-only strategy by 16%. Most of the increase was due to increased weight of purchasing, rather than the greater penetration of the market - and this in turn was attributed to a more powerful communication of the advertiser’s message.
USA: STAS of television and magazines
John Philip Jones, a professor at Syracuse University in New York who spent many years working at J Walter Thompson in London and elsewhere, has analysed and compared two sets of sales and exposure data - one dealing with television advertising and the other with magazine advertising.
His technique is to produce a summary measure called STAS (Short Term Advertising Strength) which represents gain in market share of sales. A brand's market share of sales among households not exposed to its advertising during the seven days before a purchase is called the Baseline market share (for example, 10.0% share). Its share in households exposed to the advertising is the Stimulated market share (for example, 11.5%). The Stimulated share is indexed on the Baseline share, and this index is the STAS figure (for example, 115). For a campaign, the STAS figures for each week are averaged to create a campaign STAS. It is a good measure of the effectiveness of the advertising.
For measuring the STAS of television Jones used a year’s data from A C Nielsen’s single-source Household Panel [123]. He examined 78 fast-moving consumer goods brands across 12 product fields. The result, averaged across all 78 brands, was a television STAS of 118.
To measure the STAS of magazines Jones turned in 1998 to the most extensive body of single-source data available on brand sales and magazine readership: 110,000 interviews by the Starch research company in the US during the years 1959-1964. [124]. They collected information on purchases of 73 packaged goods brands and exposure to 707 advertisements for these brands in the magazines Life and Saturday Evening Post. The outcome, averaged across all 73 brands, was a magazine STAS of 119.
The similarity of the television STAS of 118 and the magazine STAS of 119 is striking. The clear conclusion is that magazine advertising is equally as effective (per exposure) as television advertising, when each is used on its own.
Other case studies of the sales effectiveness of mixed-media schedules are given at www.hmaw.net – including Tim Tam biscuits in Australia [82] and, from JWT in USA, a packaged goods product [125].