With a general election looming, Britain's market research wizards are on tenterhooks. The latest issue of the UK's Market Research Society magazine is titled 1997: Year of Decision for MR, and the lead article starts by asking whether 1997 will be a year of 'further embarrassment'. The second article is titled: 'Laying the ghost of the nightmare of '92', a reference to the way UK opinion polls were all at sea with their predictions of the 1992 UK election result.
Having celebrated their society's 50th birthday last year with much braggadocio, researchers now have a hangover. This being election year, the opinion polls are on trial again, and if researchers get things wrong - which hardly seems possible, but then it did not seem possible last time - their credibility will be in smithereens.
Inevitably, the entire market research business will be seen as guilty by association and market research is now pretty big business. With £7bn in turnover and steady, compound growth of 8-10% a year, the UK research business has done extremely well of late. Moreover, the Association of Market Survey Organisations (Amso) forecasts growth of 15-20% this year.
Research is a nice little export earner, too. As with other marketing activities, Britain is a world leader. According to Amso, about 20% of UK market research turnover comes from international surveys and international work is growing faster than domestic.
Today there are few manufacturing companies, media houses, government departments, public bodies, academic institutions or even retailers in Britain that do not integrate research into their decision-making processes. For many years, retailers were market research Cinderellas. They believed they could learn from their store traffic everything they wanted to know. As a result, they felt they did not need to ask any questions. But retailers now account for some 5% of total research expenditure and their share is growing steadily.
However, not everything is rosy. Although other market researchers strive to distance themselves from the political pollsters, they know they share their problems, and many of those problems are fundamental. On one hand, some members of the public manifestly enjoy taking part in research surveys and focus groups. The paid interviewers know who they are and return to them repeatedly. This leads to bogus results.
On the other hand, more people are becoming refuseniks. In Britain the industry now conducts well over 15 million interviews per year and 100 million questionnaires are mailed out by database companies. Although there are no published figures, most experienced researchers will confirm that refusal rates are rising steadily.
Even in flagship studies like the National Readership Survey, response rates are now just over 60% after eight or more calls on respondents. (When the survey started in 1954 they reached 85% after six calls.) So 40% of the NRS universe is nowadays uncanvassed. Political polls suffer similarly high non-response rates, which was one of the reasons for the 1992 debacle.
More worryingly, a welter of evidence shows that a growing proportion of the public is now playing games with market researchers. They understand the processes (which are hardly rocket science) and consciously manipulate their responses. Like voters in by-elections, they deliberately send messages to companies and political parties, using the researchers as conduits. It takes only a small percentage of such respondents to invalidate results. None too soon, the research industry has started to investigate these problems. Several of the key papers at last month's Amso annual conference addressed the increasingly uncertain relationship between market researchers and their raw material - the public.
It is by no means clear what, if anything, can be done, and that is why so much hangs on the election pollsters' accuracy this time round. Polls capture the headlines. Polls have also captured politicians. Nobody pays more attention to surveys than politicians. Contrary to conventional wisdom, market research is a far more potent influence in modern politics than advertising. Politicians read, mark, learn and inwardly digest the voters' views on everything from abortion to traffic crossings. And the pollsters revel in their influence.
Behind the Oval Office, the new sell-and-tell potboiler by President Bill Clinton's former chief electoral strategist, Dick Morris, shows Clinton messily awash with market research findings in the run-up to last year's election. Indeed, Clinton often felt himself drowning in them.
The pollsters tried to control his every waking moment. They told him to become a father figure with a powerful red tie and dictated how he should spend his holidays. 'Can I golf?' the president asked the pollsters, his tone dripping sarcasm. 'Maybe if I wear a baseball cap?' He was told: 'No, sir, go rafting.'
In compensation, market research has given political leaders, both in western and ex-communist democracies, new power over their parties. In Britain, former prime minister Margaret Thatcher was able to march on with confidence during the Falklands War because she tracked public opinion throughout. Neil Kinnock, the former Labour leader, was able to defeat Militant extremists because he knew that the vast majority of the electorate, and the vast majority of Labour voters, supported him. The ability of the present Labour leader, Tony Blair, to ride roughshod over Labour's historic articles of faith has, again, been built on firmly researched foundations. When zealous and partisan supporters claim 'the party'll never buy that,' the leader, research data in hip pocket, can nowadays ignore them with insouciance.
All this emphasises the present symbiosis between market research and politics. It will last as long as the politicians continue to have faith in the researchers' wizardry. The politicians influence a great deal of business, both directly, via government departments, and indirectly, via publicity and the media.
So the researchers wait - and worry - for 1992 was not their only cock-up. They also bombed in 1970. And in the 13 elections between 1945 and 1987, some 40% of pollsters' forecasts were inaccurate by more than plus or minus 2% — a small margin, but usually sufficient to swing the result, which is all that matters.
Most of the time the researchers' findings cannot be verified. They cannot be proved right or wrong. Except at elections. That is why market researchers are now so jumpy. Their outlook is far from bleak, but it is distinctly unsettled.
Source: Fletcher8 (reprinted with permission)
Was this article helpful?
You won’t want to miss this. Get Paid Taking Surveys In Your Spare Time. I know what you’re thinking, ‘Oh great, another get rich quick scheme’. WRONG! This is your guide to making money from home by participating in paid surveys on the internet.