Image of boat and water.

ASA: blind taste test not misleading - Bulmers

Published on 09 August 2018

What information do you need to include in a comparative ad to support your claim? Can you round up figures to the nearest percentage if you make this clear in the small text?

The background

A September 2017 magazine ad for Bulmers Cider released by Heineken UK Ltd t/a Bulmers (Heineken) stated "2/3 of drinkers prefer the taste of Bulmers Original to Magners Original". The small text of the advert stated "SOURCE: Cardinal. 65.8% in a head to head blind taste test, surveyed in Nottingham and London June 2017. Excludes those who expressed no preference. Sample size: 146 regular apple cider drinkers".

The ASA received two complaints challenging whether the claim that two thirds of drinkers prefer Bulmers Original to Magners Original was misleading and could not be substantiated. Heineken stated that the claim was based on a study undertaken by Cardinal Research and argued that the sample size was robust and as such supported the extrapolated conclusion that two-thirds of drinkers preferred Bulmers Original to Magners Original.

The development

The complaints were not upheld by the ASA.

The ASA judged that the ad made clear the number of people surveyed and that readers were likely to realise the '2/3' claim was based on the 146 people surveyed, and not all drinkers. The ASA did note that 65.8% was 0.9% less than the equivalent percentage of two-thirds, but the fact that the unrounded figure had been provided in the small text meant that consumers were unlikely to be misled by the gap between the figures.

Heineken provided a copy of the study methodology to the ASA, which indicated the number, gender and age category of the participants in the study. The methodology also explained that the proportion of male and female participants in the two age categories (18-34 and 35+) were weighted in accordance with Kantar Worldpanel Alcovision data in order to ensure that the sample was weighted to give a representative sample of cider drinkers. The ASA took expert advice on the methodology and considered that for such a market research survey, the sample used by Heineken reasonably matched the Kantar data and as such was sufficiently representative of the cider drinking population. The ASA also concluded that the sample population was large enough for sufficiently sensitive results to show that around two-thirds of people preferred the taste of Bulmers Original to Magners Original.

Considering the above, the ASA considered that the sample size and methodology were adequate to substantiate the claim "2/3 of drinkers prefer the taste of Bulmers Original to Magners Original" and therefore the ad was not misleading.

Why is this important?

For companies considering launching a comparative advertising campaign, this ASA ruling provides useful guidance to help ensure that such an ad does not fall foul of the CAP Code. The ruling helps clarify what information must be included to support a claim, and also what the ASA will consider to be a 'reasonable' study in order to substantiate such a claim.  

The discussion of the discrepancy between 65.8% and the percentage representation of two-thirds also indicates that the ASA is willing to take a pragmatic approach to such discrepancies, provided the small text draws attention to the true number.

Any practical tips?

Comparative claims are ripe for attack from consumers, competitors and statisticians (!) alike. Ensuring robust evidence and methodology is behind a comparative study is critical in order to substantiate the data, noting that a copy of the methodology and details of any studies may be needed for passing to the ASA to defend a complaint. The small text of the ad must also include enough information to inform a consumer about the number of study participants, the conditions of any test, and any other material information. Finally, even if there is only a slight difference between any figure claimed in an ad and the true figure, ensure that the small text includes the actual figure.