Will on-line make us better researchers?

By Piers Lee, Managing Director of Kadence International Pte Ltd and President of the Market Research Society of Singapore

Asia Research completed its 3rd annual review of the on-line research business in Asia.  Such ‘reviews’ are in reality a window into the on-line research business for the next 12 months, highlighting new technologies, trends, and how this will impact on the broader market research industry in Asia.

On-line research and all its associated technologies and support industries is now the fastest moving part of the research industry in Asia.  With clients now a bit tired of so-called ‘new products’ from the agencies (often just repackaged methodologies from the past 10 years) some of the factors that will bring value to the industry such as access to difficult-to-reach audiences, interactive research, and speed can be met through on-line research.

Based on interviews with Ludovic Milet, MD of GMI Asia Pacific; Martin Filz, MD of Research Now Asia Pacific; Chris Rowen, CEO of i-Link; and Jason Smith, President of Innovation from the proprietary panel company VisionCritical, Asia Research reports on how on-line research is likely to change the landscape for research in Asia, and specifically for people working in the main stream agencies.

Firstly, one of the key differences in Asia is how developed and emerging markets sit side-by-side with each other.  In Singapore, on-line research accounted for 20% of all research methods (according to ESOMAR 2009 report).  Walk across the Causeway into Malaysia and it drops to 3%.  Mostly these disparities are a function of broadband internet penetration within the market.  Ludovik Milet from GMI comments “in most emerging markets broadband internet is still quite expensive relative to household incomes, and hence on-line research in markets like Indonesia, Vietnam, and the Philippines will be for niche projects conducted among trend setters, for example for new product development research”.

Martin Filz from Research Now also points out “economics has a lot to do with lower penetration rates of internet research in emerging markets.  Traditional data collection methods, e.g. face-to-face, still have quite low unit costs in markets like Indonesia and the Philippines.  In other markets investment in existing infrastructure such as CATI centres makes some research companies reluctant to use on-line research, although the progressive move to outsource fieldwork means that on-line research will continue to grow even in the more mature markets”.

Martin observes that in the most developed markets for on-line research including Canada, Australia, and Japan, on-line makes up 35-40% of all quantitative methods.  So markets like Singapore and Korea where on-line accounting for less than half of this, still have considerable room to grow.  One area for potential growth is migrating brand trackers from traditional in-person or CATI methods to on-line.  One of the benefits is that many brands in Asia are known more by familiarity with logos rather than their phonetic name, and therefore these surveys lend themselves better to on-line (visual) research methods rather than by telephone.

It’s a question of suitability:

The growth of on-line research in Asia has so far been driven mainly by economics, important in the last couple of years of corporate austerity.  However, the panel companies emphasize that this is the wrong motive to use on-line research since many surveys that agencies wish to move to on-line are unsuitable.  The long, ‘boring’ surveys are quite hard to migrate to on-line because this leads to high dropout rates among panelists.

However agencies, that are under continual pricing pressure keep looking to on-line research as their savior and this brings into question quality issues.  Recent bad press over on-line survey quality, for example the recent ‘undercover article’ in Research Live has made clients nervous.  Martin Fliz from  Research Now points out that quality issues affects all data collection methods, whether it is interviewer bias or poor recruitment for focus groups.  Martin points out that on-line panel companies are doing far more than research agencies to ensure fieldwork quality.  Since panel companies specialize in operations and project management, and do not get involved in analysis or insight they can devote far more resources towards quality control.  These include measures such as PC finger printing (ensuring the same PC is not used twice in the same survey), analysis of IP addresses, length of surveys, logic tests, ‘click tests’ (ensuring that respondents don’t give the same answer to scaled questions or in grids), or grid speeding.  Industry Standards such as ISO26362 (panel standards) and ESOMAR 26 are also helping to set the proper standards.

Ludovic Milet says that quality is now the biggest differentiator between panel companies.  “There has been an explosion in the number of on-line panel companies in Asia in the last few years.  However the size of panels and the price of on-line surveys are no longer the selling points.  Instead panel quality, speed, and client servicing are the key differentiators of the good panel companies.”

But one of the best quality control measures is in the survey design itself.  If surveys are interesting, punchy, and relevant, respondents will want to take part not just for the financial incentive but because they want to make a contribution.  Part of this process has been to make the presentation of the survey better for respondents.  Chris Rowen from i-Link says that the on-line business is effectively two businesses, 1) on-line access panels, and 2) on-line field services.  On-line field service companies like i-Link have invested in panel software that is adaptable and makes the survey far more visually pleasing to the respondent.  When faced with an enormous grid of 20 questions x 10 point scale responses, the consumers’ mind might just shut down.  By redesigning these into more user-friendly, well presented surveys using flash technology it helps to keep the respondent engaged in the survey.

Cutting out the middle man?

Most panel companies (but not all) have avoided the temptation to set up their own insight divisions and most still rely on the ‘traditional MR agency’ for 70% or more of their business.  However, end clients are increasingly going direct to on-line panel companies for their survey needs.  Corporations now have customer feedback from many more sources, not just from their traditional MR agencies.  Corporate websites, blogs, bulletin boards, and social media are all channels of customer dialogue and in response to the developments in new media corporations are setting up their own analytics divisions and implementing ‘Enterprise Feedback Management’, a system of controlling the dissemination of information to stakeholders in their business including customer feedback.

The Asia Research Buyer Survey conducted in June 2010 showed that about a third of corporations in Singapore are buying surveys directly from on-line panel companies.  Chris Rowen from i-Link says that some corporations buy (on-line) fieldwork direct from panel companies but still involve the MR agency to undertake the insight and reporting.  Both GMI and Research Now say they are approached by end clients directly for smaller, tactical, fast turnaround surveys for example dip stick brand awareness or a pre and post adverting survey involving visuals.

So what’s next?

It’s not difficult to talk to on-line panel companies about the future!  On-line research is following consumer lifestyles much closer than the traditional research industry and of course this is changing fast with technology.  The main area of change is that of mobile technology, and this has even higher relevance in Asia.  While internet penetration in many markets in Asia is low, penetration of mobile phones is far more on par with developed markets (see Tapestry Works article on Page #).  Chris Rowen from i-Link comments “call up a household seeking a young male for a telephone interview, and they won’t be in!  You are far more likely to reach them on their mobile.  From a research design point of view, the move from CATI to on-line was not huge, with companies generally using the same kind of questionnaire layout, even if this is not always the most desirable!  However, from a design point-of-view the move from on-line to mobile will be a much bigger jump, and researchers will need to be trained in how to make this jump.”

Research Apps are a way forward, but currently there is no industry standard.  This might also represent a potential breakthrough in one of the final frontiers of on-line research – B2B on-line access panels.  Many business decision makers have smart phones and while they are often not at the end of a landline they might find themselves with 5 or 10 minutes to spare at airports, in lobbies and taxis, etc.  These are opportunities to express their views on airlines, hotels, corporate banks, and even mobile phones!  But the panel companies still recognize that widespread use of B2B on-line research is still far away, a consistent finding from the last three years of Asia Research reviewing the on-line research business.

On-line research will also change the way end clients procure research.  A significant part of a client’s research budget is spent on regional brand trackers.  A lot of these trackers provide key metrics to their board of directors on brand health; clients will be able to save a significant amount of their research budget in implementing trackers directly with on-line panel companies freeing up more budget to spend on higher level research, or more complex ad-hoc research requiring more value-add insight.

More immediately though, clients are keen to tap what they view as one of the richest sources of consumer feedback in the world today – the social media networks, e.g. tapping into the 500 million users of Facebook.  New tools are being developed to try and weed out the ‘noise and junk’ in social media, to bring the relevant information into the clients laboratory for further analysis.  Organizations such as buzztrend.com trawl notice and bulletin boards for this information.

There has been much debate about the usefulness of this type of research.  The main issue is whether good or bad press about brands in social media is relevant when you have no idea of who is saying what.  Jason Smith from VisionCritical comments that “the sampling issue with social media research has not yet been nailed.  The CMO of Coke might like it (social media research) since it gives an indication of the overall health of the brand, but the Product Manager will not since it does not inform on who is saying what and the chatter is usually not specific enough to individual products”.

Chris Rowen from i-Link comments that social media usually captures either advocates or adversarials and not enough of the ‘typical’ consumer.  But like other on-line research it still has its place in the family of research methods but should be treated more as qualitative research.

However one factor that might bring doubt to social media research (which has even got its own name now – “web raiding”) is the deliberate ‘talking up’ of a brand on the Internet, e.g. coming from the client’s own marketing communications department or sales people.  Or deliberately talking down of a brand or company – nicely out of the reach from the defamation law suits and libel lawyers!  A cyber war of words between brands could be a feature of the next decade.

So what?

On-line research is facilitating what will probably be the biggest shift in the way market research is conducted since the birth of the industry.  The biggest shift is about consumers and businesses ‘opting into research’ rather than being approached by an interviewer and agreeing sometime reluctantly to a question and answer session.  Even the reliability of these answers in terms of predicting future consumer behavior has been brought into question.

Inevitably more research will be about observing, e.g. brand chatter on the Internet, and incorporating research into the everyday activities of consumers and businesses.  In some cases research needs to be used when the ‘respondent’ is not even aware of being researched, e.g. observing the way they use media; deeper analysis of their response to advertising; how they behave during the sales cycle such as what questions they ask and what objections they raise; how they navigate their way around websites or traditional sales channels such as markets or malls.

Research surveys might be incorporated into computer games, where real consumer choices in a purchase path become part of the game whether this is in your weekly shop, or the decisions entrepreneurs make in setting up a new business, e.g. choice of office location, banks, IT brands, staff, stationery suppliers, airlines, hotels, etc.

So will on-line research make us better researchers?  The answer is ‘yes’.  The key reasons will be:

  1. Better survey design – by relying more on ‘opt in research’, surveys will need to be more relevant, punchy, user-friendly and interesting.
  2. More insight – if panel companies undertake the fieldwork and project management it is then down to the MR agency to demonstrate the insight.  If MR agencies cannot do this, clients will no longer use them!
  3. Opening up of markets – clients have traditionally placed their regional brand trackers with the larger agencies with the networks throughout Asia.  If the on-line panel companies provide the boutique agencies the regional resources for data collection and project management, clients might seek new avenues to source research for brand health.

But don’t ask me, I could could be a biased sample!