Market research on a regional scale, once the accepted route for companies wanting to test new products or ads on the market before their launch, is now under scrutiny. Marketers are questioning its validity for predicting how the product will perform on a nationwide scale.
Regional market research findings can be deceptive as different parts of the UK have their own idiosyncrasies, making them likely to give inaccurate indications of how the product will perform nationally. Justifying regional research is becoming more difficult as, on the one hand, companies are spending exorbitant amounts of money on testing while, on the other, those tests are producing less reliable predictions.
One of the main reasons for this is that the UK regions are becoming increasingly demographically diverse. According to FDS International chairman and chief executive Janet Weitz, the difficulties companies face with regional testing are compounded by the fact that the UK market is becoming increasingly fragmented. “Years ago standard market research would be London, the North, the West, Yorkshire and Scotland. But it’s not that clear-cut any more,” she says. “Many things have driven fragmentation, not least of which is the increase in ethnic areas in UK regions. In Birmingham, for example, you have a lot of different ethnic communities, so areas need to be broken down even more.
“It doesn’t take a genius to work out that mass-marketing, which works so well in middle America, does not work here.”
Weitz says that only the youth market lends itself to simple regional research. Even so, she still believes companies would be wise to test their market in a number of areas.
“There’s not much difference between the 18- to 25-year-olds in London and those in Birmingham, for example. But if you’re going to test the market, you have to do so in more than one area,” she says.
Anne Thompson, a divisional head at research firm Market Profiles, agrees with Weitz about the regional disparities that exist in the UK. She argues that these differences must be built into the testing procedure if market research is going to have any relevance.
She says: “The regional differences are something you have to build into every project design as they can vary immensely according to the product. But it shouldn’t be a major problem. Providing the design is right you can still get a good national prediction.
“Look at the demographic profiles of the various regions and aim not to select the extremes; however don’t pick regions that are too similar either. You need to be able to identify the differences between the regions and compare them with the national population.”
But are there any regions which provide the best test environment for UK-wide predictions?
Ideal testing ground
Weitz argues that there isn’t any particular region of the UK which provides ideal testing conditions. She says the only way for companies to gain a reliable picture of the UK is to randomly sample as many regions as is financially possible.
Thompson, like Weitz, also highlights the financial aspect of market research as a limiting factor. She says it is something of a curate’s egg in that to get the best reflection of the UK market you have to sample as many regions as possible, but covering a lot of regions is very costly. “I don’t think there’s a region that’s best for testing,” she says. “Depending on the budget, look at how many regions you can cover to give you the most complete answer.”
However, Roger Parkyn, chairman at advertising agency CKMP believes Ireland provides a good launchpad for any client interested in the UK market. He says: “Ireland is the perfect test opportunity because you don’t get the overlap you tend to get between TV regions in the UK. It’s very self-contained. Also, the cost of being on TV in Ireland is a small fraction of the cost in the UK, so you can map out some serious TV advertising in Ireland without it being a huge investment.
“And Dublin is very much a European city; it’s very cosmopolitan, with a lot of companies in developing industries based there,” he says.
Parkyn points to the fact that brands such as Procter & Gamble’s Pantene, Wrigley’s Airwaves and Cadbury’s Twirl were all tested in Ireland prior to their full UK distribution. “You need only look at what the big multinationals are doing. Cadbury, Heinz and Nestlé already have their Irish operations in place,” he says.
CKMP’s clients might typically be smaller UK companies which cannot afford a national launch; US companies which have yet to develop a European marketing operation; and European companies with niche brands which may want to expand them into European brands. “If a client wants to develop its brand, then we would advise Ireland as the best place to test it out,” says Parkyn.
Parkyn says Ireland is also ideal for small manufacturers wanting a wide retail distribution to test their products. In the UK, he says, it is difficult to achieve a wide distribution which will cover small independent retailers to all the supermarket chains in particular regions.
In Ireland, however, this is possible, and it’s a small enough region to keep the cost of TV advertising low. Parkyn says: “You can have a substantial TV campaign in Ireland for about &£500,000. In the UK, the same would be a major expense – and the regions are too idiosyncratic.”
While Ireland clearly has its merits, Larry Rees, director of strategy at advertising agency Underground, argues that the regions used for research should vary according to each test case. And technology is enabling it to develop different research models for different clients.
Underground will first determine the client’s objectives – what brand value they are aiming to create in sales terms – and from that it will calculate the investment the client needs to make. Data is then collected from sources such as the TGI and BARB which provide information on behavioural patterns on a regional basis.
“The old way was to look at that information and use it to guide your thinking,” says Rees. “Now, we can calculate the different costs of advertising in different parts of the UK, particularly TV costs that can fluctuate enormously.”
Rees says they aim to find two or three regions that bring together the attributes of a client’s target audience, while also behaving in a way that reflects other parts of the country. Ideally, the regions would also sit on the lower cost side for mounting a test.
But Rees adds: “There’s no right answer to how you should mount a regional test. We tend to run a number of simulations to present to the client, with a reference point of what we believe will happen to the cost per sale of the advertising based on different scenarios.
“We have never run a test using the same mix for different clients. Each one has been a bespoke solution. That’s the beauty of computers; you’re now in a position to scientifically model the test parameters that you’re looking at. You won’t get a perfect answer but you’ll reduce the risk. “
Rees maintains that there is no one specific region that offers the ideal testing ground. Research will inevitably vary according to the product or ad that’s being tested and the financial limitations of the client.
For a market test to be a effective as possible, companies should sample as many regions as possible to account for the UK’s demographic diversity. But, as Rees points out, the ideal research model is likely to be a bespoke one, created according to both the client’s needs and limitations.