Eight out of ten cat owners once said their pet feline preferred to eat Whiskas. It remains a mystery how this feedback was obtained from the animals in question, but the statistic has become one of the most famous advertising slogans in the UK.
This campaign also demonstrates the power of using research in advertising. This can also be seen in the latest ads for GSK's mouthwash brand Corsodyl, which uses bold headlines such as '79% of over 35 year olds suffer from gum disease' as part of its 'Campaign for Healthy Gums' activity.
This is an arresting statistic. However, marketers planning to use such information in their advertising and PR need to be careful; consumers can go through data with a fine-toothed comb and decide for themselves whether it stands up.
There is also the risk of information fatigue. Journalists receive so much research from brands - most of it self-serving - that the headline findings must have immediate cut-through in order to receive coverage. This need to be noticed by journalists encourages exaggeration of the findings.
Mark Hodson, founding partner at Opinium Research, says: 'So much research is done to give a business a competitive advantage, but you have to be very careful because people are much better at seeing through numbers now.'
Auction website eBay fell foul of the Advertising Standards Authority (ASA) last year following an ad that claimed its prices were '25% cheaper than the high street on brand new items'. The claim was based on independent research on the average price of 288 products, with recent eBay transactions compared with prices in mainstream retail stores, such as Debenhams and John Lewis.
The ASA banned the ad on the grounds that it was misleading because the average figure might disguise the fact that one high-street store was regularly cheaper than eBay.
Lucy Davison, managing director of PR agency Keen as Mustard, and a former global marketing director at Research International, handles PR for market research agencies. She suggests that campaigns must be carefully planned before research is commissioned. However, this must not come at the expense of being fair and representative.
'You can't necessarily guarantee that you will prove something,' she says. 'You go in with a hypothesis. If you ask the right people, you will get the right result, but that could be misleading. If you ask a stupid question, you get a stupid answer.'
Davison warns against using 'easy-to-do' online research tools, such as Survey Monkey, for high-profile campaigns. She adds: 'You can actually get the research done quickly, but in reality you have to be prepared to do the work.'
At a stretch
Kate Thompson, partner and head of B2B Research at agency McCallum Layton, suggests the biggest danger when injecting statistics directly into campaigns is over-stretching. She cites a piece of research that gauged London businesses' view of the recession. 'There was a temptation to stretch those findings and to comment on them from a wider perspective, but that wouldn't have been right because it was only the London view,' she adds.
Even if a brand's research is robust and compelling, it still needs to be presented in a way that is easy to digest. Nothing is more likely to put someone off than a confusing graph or long list of numbers. Hodson says that consumers are becoming more numerate but simplicity is still order of the day.
'You communicate best when you use a graphic rather than a load of figures,' he adds. 'When you can turn a number into a memorable phrase, that is the secret.'
However, research does not always deliver, especially when it misses the point of the brand in question. Hodson recalls the studies that preceded the launch of the first Sony Walkman, which, he says, concluded that there was no point in having a tape player that could not record, had no speakers and so required the use of headphones.
'Clearly, the research started too close to the physical concept,' he says. 'It didn't establish the need and solution, which the device could offer.'
The final word on the subject goes to comedian Vic Reeves, who famously said '88.2% of statistics are made up on the spot' - an insight subsequently used in an ad campaign for Guinness.
While Reeves' tongue was firmly planted in his cheek, the sentiment prevails; there is a lot of cynicism around research, so do not misrepresent the statistics.
How to use research in a campaign
DO plan ahead - get research done long before the campaign is due to break, so there is time to react if it comes up with unexpected findings.
DO accentuate the important findings - the statistics that will stop people in their tracks.
DO use research to add weight to existing communications and established truths.
DON'T try to support a statement that is ultimately untrue.
DON'T twist the statistics.
DON'T use obvious, self-serving statements; for example 'soft drinks are good for your child'.