“The group or agency that plays the data game very well – how you manipulate data, how you analyze data, and how you bring it to the client in a smart way – will win the battle long-term,” says Elie Khouri, regional managing director of Omnicom Media Group buying and planning agency. “It will have a greater share of client business. And I think that’s what everybody’s missing – the importance of data and the importance of consumer knowledge.”
In the wild world of advertising, data dominates. Roaming the wilderness without it can throw up trap after trap, with agencies falling prey to misinformation.
Any agencies, planners, or buyers worth their salt know the rules of the game: pre-plan, plan, and post-plan, using data all the while. But sorting through research and trusting its accuracy are not necessarily easy things to do. There’s a lot of data out there, and its volume is increasing.
“[Data includes] any information, any research, [including] any analysis that you do by going online and looking at consumer trends and looking at tweets, looking at Facebook, looking at social networks, analyzing all of this complexity that we have today,” says Khouri.
To get a high-quality picture of consumers, marketers must navigate these reams of data, whether it’s about consumer trends, consumer habits, television viewership, print auditing, online polls, opinion blogs, radio audiences, or even psychodynamic motivational qualitative research. This volume of available information can become a problem for those on both the agency side and research side alike.
Planners and researchers say that the problems afflicting research are myriad, but not untreatable. “The biggest challenge that we face is the definition of the [research] problem itself,” says Aruni Ghosh, research manager at Nielsen. “Clients, local or multinational, request research if they have a marketing problem, for example. If the problem is not clearly defined, the output you get may not be addressing the problem. If you leave it open ended, it might sound convenient for us, but it’s not. Tell me what action you’re going to take based on the research and give me a good research brief, and my recommendation will be much more focused.”
QUESTION TIME. Once research agencies have zeroed in on the right questions to ask in order to address a well-defined problem, they’ve got to get people – generally members of the public – to respond to their queries. George Khalaf, client service manager and research analyst at audience measurement company PARC (Pan Arab Research Centre), believes the main problem that researchers face is getting information in the first place.
“The challenge is rejection mainly,” he says. “Especially face-to-face. Even when you call people, they will say no to answering a survey and hang up. Once you interfere in people’s lives, you get a high number of rejections.”
Khalaf finds that people respond to such “intrusions” differently, and this issue makes it difficult to build an inclusive sample. “We target different demographics, and we found that Westerners refuse the most, especially when you probe into their lives, and a lot of them don’t have much time. But we try to make the sample comprehensive, to cover all demographics.”
Elie Aoun, COO at Ipsos MediaCT MENA, says that the challenges that researchers face differ from region to region and country to country, but that in the Middle East social barriers make it difficult for research companies to undertake their work.
“Face-to-face interviews, especially in homes, are difficult to do, so that’s why we have to find creative solutions to reach those people we want to interview,” says Aoun. “For example, telephone interviews are much more accepted than face-to-face interviews in our region, and we are now introducing online as a way to collect information from people.”
And the way a question is asked is extremely important, according to Nielsen’s Ghosh. “In market research we get extensively trained on how to ask questions of consumers,” he says. “Many reports which you see in papers or magazines shouldn’t be taken very seriously, because the way questions are asked are often very biased. For example, if you ask, “Do you regularly buy health insurance when you travel?” there is no apparent mistake in the question. But if I look at it closely, how people define “regularly” can be very different; for me it could be every month, for others it could be six months. The technical part is very important.”
QUALITY CALLBACK. The experience and skill levels of field researchers vary from entry-level students to seasoned professionals coming from other research agencies. In order to streamline the questioning process, Ghosh says that researchers in the field need to be extensively briefed on how they should be asking questions. In addition, even after the questionnaires are submitted back to the research agency, they go through a separate quality control center.
“We typically do this checking by calling 30 percent of the sample size back to be verified,” says Ghosh. “If there’s a discrepancy, we go back and re-do the questionnaire.”
Polling a fair representation of research subjects is of prime importance to all research agencies and media professionals looking to use the data gathered, but Georges Naaman, general manager of media agency MPG, says sample size is problematic in the UAE.
“The methodology of creating a sample size should be based on official census numbers which are given by the government,” he explains. “The last official survey is quite old [2005], so there are no official numbers – neither are there in Saudi Arabia. Any sample size should represent the variety of the population in terms of demographic information.”
The lack of census information leaves a large margin for error, particularly in the UAE, which boasts a significant transient element in the shape of expats.
Khalaf from PARC says that there is always room for error when sampling a population. “With statistics there is always a level of error,” he says. “The best thing to do is to make sure the sample is as representative as possible.”
NEW NUMBERS. Despite the challenges, agencies and other interested parties are by no means at a loss for data. They have plenty of numbers to crunch and analytical work to do. And the rise of digital and social media has opened up a whole new avenue of potential research. Naaman, who worked with Ipsos in Beirut as well as other research companies before he came to MPG, says that research companies must revise and adjust their methodologies to cope with this evolution of the market. He explains that with the high-tech invasion of new media contact points such as computers, mobile phones, and games, people’s habits are changing.
“Those technologies are now part of our daily lives,” he says. “All advertisers are aware that the spend behind these contact points is increasingly high, but that’s happening without proper planning due to the fact that we’re lacking proper surveys. What we need are proper online consumption surveys, and surveys that monitor spend on Web sites, which are available in more advanced markets such as Europe and the US. In the GCC, Ipsos and Nielsen are currently working on developing such projects, but they’re still in talks.”
Ziad Skaff, internal group director at the insight and research agency of OMG, Integral, says the challenges that face researchers have indeed evolved today due to the abundance of information available, but the techniques of data collection haven’t evolved to keep up with the times. “We try to work as much as possible with different suppliers; when you don’t trust one source completely, it’s better not to put all your eggs in one basket,” he says. “We make sure as much as possible that the data coming to us directly is at a raw level, so we can validate it internally in terms of sample size, distribution, and skewing, before adapting it,” says Skaff, whose company outsources data collection.
With so many information sources to choose from and so much data to sort through, it’s important for anyone using research to select their sources with care and consideration, thus making sure that the data they use before, during and after campaigns is as robust as possible.
“There’s a lot of information, articles, and data out there, and everyone can claim to have done research,” says Integral’s Skaff. “Sometimes it’s very poor quality data, and I’ve seen that a lot. I’m not saying that all the output coming from suppliers is wrong, you just need to be really careful in how you pick and choose your studies.”
So, regardless of how marketers choose to interpret or use the information they gather, one thing is clear – the quality of that data depends first and foremost to the quality of the effort expended in collecting it. And marketers that make that effort will take a significant strategic advantage into the data game.