John Griffiths, Planning Above and Beyond, reports on the second day of Online Research 2008, where major themes included creating effective online communities, market research in China, data quality control and how to effectively mix research methods.

Collaboration, co-creation and community

Mike Cooke introduced the second day with a panel of four making predictions about the next year. Collaboration, co-creation and community were the watchwords. By letting the greyhounds out of the traps so close together, it was inevitable that ‘convergence on steroids’ became more of a watchword than it was supposed to be, and priorities started to merge.

Tantalizingly, Louise Ainsworth, of Nielsen, referenced Web 3.0: the seamless semantic web, working very differently from the social media of today. Luke Allen, of Nunwood, talked about the growth of complex interactive environments more mainstream and practical than Second Life.

Nick Blunden , of Profero, had obviously realised the dangers, and produced five key words including, paradoxically, a polarisation towards anonymity and intimacy. The last speaker, Andrew Walmesley, of I-level, thus had much to do to lift himself above the semantic web that the previous speakers had woven. And so he did.

He pointed out that if Web 2.0 is about interactivity between people, then the role of research has changed, and brands need to play a listening role, rather than using research as a command and control feedback mechanism. This was brilliant and radical – and almost certainly miles ahead of most client practice.

By showing how differently research needs to be framed to be insightful, he thus posed a very interesting question we should all consider: How am I doing? His other bon mot was that brands needed to learn how to be a good date (and by implication not pickup artists).

Online surveys and technology

The second session shifted us away from the bells and whistles Web 2.0 brigade with a bump. In a low key presentation, Charles Pearson, of Research Now, made a careful comparison between the results of Flash surveys using graphical tools and those using more conventional text and buttons. Research Now has opted for Flash as their default platform for questionnaires design, and the main metric seemed to be completion time – which is problematic because time isn’t a great indicator of engagement.

The ambiguity of images can also lead as much to research fatigue as banks of words and phrases. The presentation focussed on PC-based surveys; what Pearson didn’t draw attention to was that fact that everything that was true of PCs was also applicable to the mobile platform. His was one of the most interesting and useful presentations of the day, and Adobe would give their right arm for information like this.

Dr Tacis Gavoyannis talked about young people and their use of technology. The most interesting observation was the offset between the internet channels used by teenagers and those used by their parents, and the extent to which this was driven by teenagers with time and the savviness to get it to work, and the extent to which it was the result of the need to stay out of sight of parental supervision.

Serving the Snackrs

Then came a two-hander between Jo Carter of United Biscuits and Graeme Lawrence of Virtual Surveys. Considering how few clients have got around to experimenting with research communities (if the testimony of the clients presenting case studies in the casestudy café is anything to go by) we owe a debt to Virtual Surveys, which not only persuades its clients to do this kind of research, but to get on conference platforms and talk about it (I count five in the last 12 months).

Some of the material was familiar: user typologies of rabbits, tortoises and dodos, for example. What was interesting about the Snackrs community was how it was possible to build communities across brands for an area of common interest – potentially making it a going concern for clients who would baulk at setting up a community for just one brand.

This was also the way in which community members were being encouraged to put forward ideas of their own producing a stream of suggestions about new flavours and formats. What I haven’t yet work out is how they manage the process of refreshing the community without causing upset by firing participants, many of whom will have made a substantial contribution of enthusiasm and ideas.

The challenge of China

The last session of the morning featured John Pawle, of QIQ, and Dr Tacis Gavoyannis, for his second presentation of the day, talking about the challenges presented by China as a market. It is hard to convey much of an impression of what was said – the numbers involved are staggering – perhaps a gawp should be used as the collective term for a number of presentations about China.

One statistic I recall is that China had to opt for mobile phones because there wasn’t enough copper wire in the world for a terrestrial network! It was a shame there wasn’t more time to debate the implications of the “Sinofication” of research and marketing as services and brands get traction in what is an alien culture for Europeans, but is providing the product infrastructure for Europe and will be providing much of the service infrastructure within a generation.

Mobile markets and TV

After lunch, the clients were back with their case studies in an interesting succession of presentations. Rosica Bojkova, of T-Mobile, talked about researching across European markets, including the emerging Eastern European states. Andy More, Customer Insight Director of Vodafone, then discussed his company’s use of online and mobiles for conducting research, which often feel restricted because the penetration levels of access skew results too much.

He also provided the memorable phrase that research had to move from a recall paradigm to an experience paradigm, and that mobiles would one day be able to deliver this. The comparisons between these two presentations made for interesting listening.

Simon Scholes finished the session with an explanation of how using an online panel had become central to ITV’s fight back to get to the top of the ratings. It was clear that, once the panel was in place, it could be used in a number of different ways, including checking storylines, not only for their own shows but for those of competitors.

There were a few slides, with an apologetic aside in the notes pages for not being able to share more because of competitive sensitivities (which was hardly surprising considering there was a solid BBC delegation in the audience), but he did provide a reminder that for the most part that clients and agencies were doing their level best to share information.

Data quality control

This was in marked contrast to the session which followed on the subject of data ality. Efraim Ribeiro, from the ARF Quality Control Council, outlined the attempts the US online panel industry has been making to get its house in order. The prevailing business culture in the US allowed research agencies to operate behind high fences, pleading proprietary techniques but often sharing the same suppliers of email addresses. The result brought the industry to the edge of catastrophe as clients, most notably P&G, lost confidence in data integrity. 

Efraim’s insider account of the recovery process and the complex negotiations between a core of industry players with a circle of clients (including Wal-Mart, Bank of America and Coca-Cola) was fascinating. Inevitably, the audience was left trying to fill the gaps for areas he was unable to talk about. But it was a salutary warning to the UK industry of the risks of losing client confidence in data quality and the benefits of keep open communication channels over issues of data quality and emerging methodology as evidenced throughout the conference.

Nick Sparrow, of ICM, spoke next. A sometime vociferous critic of online panels, this was the first time he had spoken about how ICM had started to use the methodology. His remedy for the ills of panels was simple: treat panelists as employees, and police them accordingly. Fraud and moonlighting should be policed and punished, while keeping short accounts with panel members and paying them a fair wage would go far to address the underlying causes of fraud (low wages and inadequate quality control).

Millward Brown takes on the Titanic

The next presentation, by Duncan Southgate of Millward Brown, was something of a surprise. In the face of the bells and whistles brigade's murmurings about community, co-creation and collaboration, the Millward Brown approach to online research was blunt and to the point: Wanna show TV ads online? We’ll track it versus other offline media.

This was far and away the most retro presentation of the conference – and why not? I’m sure there are plenty of clients who want additional coverage for TV films for shrinking audiences. But the price of this was that the approach seem firmly “old media”, taking no account of the possibility that people browsing video online in return for watching 20–30 second ads might be in a totally different modality from watching TV commercial on a TV with others in a living room. Well done Millward Brown for staring the wishy-washy-webby-twoeys down! Time will show whether this bravura performance was worthy of an icebreaker or the Titanic.

Mixing research methods

The last presentation of the day was quite low key, and the delegates were distinctly flagging by now. David Day, of Lightspeed, provided the necessary presentation about mixing research methods between offline, online and mobile, why it is important and how to do it. There were some interesting applications of the use of mobiles for taking photos so there was a record of the context in which communications was seen.

My only quibble was that he advocated the use of SMS and WAP as the central platforms for mobile, which seemed an odd choice since Javascript and Flash functionality has been available on mobiles for years now, and offer significant benefits for both survey makers and survey takers over those lower grade formats.

Modernisers, migrators and paymasters

And so to wrap up time. Mike Cooke gathered together the threads, quoting Walmseley’s listening and dating paradigm, the likely influence of the powerhouse of China, and the shift towards the mobile as the technology that allows us to capture experiences as people have them.

For me, the two-day conference settled into a three-cornered debate between the Web 2.0 enthusiasts, the more mundane providers of online research (much of which mirrored the offline forms of research they were designed to replace), and, lastly, the clients plugging on with the migration.

It was excellent that so many are prepared to come for the whole two days. There really was that much to talk about – it couldn’t have been done in one. And given the variety of agendas the interplay between the modernisers, the migrators and the paymasters, it was fascinating to watch.

Comments on this post:

John Griffiths, Planning Above and Beyond, reports on the first day of Online Research 2008, discussing, among other things, why marketers need to look to the long tail, how Web 2.0 resembles a Swiss army knife, and if "mashing" panels could point the way forward for internet research.

Listen to a podcast
live from day one of
Online Research 2008

Mike Cooke, conference chair, kicked the day off with a set up piece that was full of quotable phrases: Marketers sell to the long tail rather than the mass (and need research to match) and that consumer conversations are viral instead of private, meaning we are digital immigrants trying to make sense of a landscape which will be home to the next generation of digital natives.

He was suitably thought-provoking, and I for one was relieved that we weren’t subjected to another “Web 2.0 is very important (so I’ll say it again)” type intro. I had been anticipating fatigue from several buzz phrases including “Web 2.0”, “Research 2.0” and “online community” (and I wasn’t to be disappointed), but this conference steered away from clichés at the start. Good.

Bill Thompson gave the keynote presentation. He was billed as a wired maverick, and so he proved. He described Web 2.0 as a Swiss army knife, and was at his most compelling when he started to itemise the components of his online persona, and the extent to which it now dominates his offline persona – he said, for example, that it was pointless to remember whether a particular conversation had occurred by  phone, email, Twitter or blog.

It was a bravura performance. His most telling comment was his expectation that anyone who dealt with him (including researchers) would have to keep pace with him through his various channels, otherwise he wouldn’t bother. So it doesn’t sound as if anyone will be signing him up for an online panel anytime soon. It also showed the extent to which research techniques need to embrace complexity and serial personas, and that a failure to do so merits instant disqualification. 

One of the questions he took at the end produced the intriguing suggestion that respondents who weren’t happy with research should be given a meta channel to discuss the issue in the terms they were happy to.  I don’t expect any Dopplr/Flickr/Facebook/Twitter/Second Life hybrid field plans anytime soon, but Bill showed where the bar needs to be set.

Ciao to Second Life

Richard Thornton, of Ciao Surveys, took the floor next. He had a tough act to follow, but got the conference’s attention by claiming that online panels were an endangered species, but that Ciao Surveys had found a way to plug the hole with Real Time Sampling.  It looked like hara-kiri, which was why he got away with it. After all, dealing with the bread and butter of online research doesn’t sound half as exciting as speculating about how to embed questionnaires into Twitter. 

Ciao Surveys’ new business model apparently lets it off the hook, as it doesn’t have to pay for uncompleted questionnaires anymore (though having said that, there appeared to be an awful lot of them). But what the company has done is to randomise the matching of panelists to questionnaires to discourage fraudsters, embedded the questionnaires into the third party websites and adopted simple demographic screeners which ought on the surface to have less wear and tear on rejected survey takers.  And providing better interviews when the right people are matched the right surveys.

It was a hard message to deliver, but clients are increasingly concerned about data quality, so someone had to take it on the chin and do something about it. It was good to get to this early in the day while people were still fresh enough to take it in.

This was followed a trip to Farside – or, more precisely, Second Life, and two very different approaches from two agencies: Repères, a French agency carrying out qual projects on Second Life, and Opinions Online, which is hustling surveys for Linden dollars.

Through the day, there were several mentions of researchers’ avatars being beaten up on Second Life (apparently this is now been declared criminal behaviour by the courts in Brussels – so THAT’S what human rights is all about), so even on the frontiers of the web it appears that people don’t take kindly to being asked their opinions in old world ways.

By contrast, the Repères people seemed to have found an intriguing activity for bored Second Lifers who could stroll through galleries or even construct objects as part of creative development. My favourite was the client viewing in the guise of a cat walking around.  I just couldn’t have stood to find clients sitting behind virtual mirrors.

Emotions, engagement and blogs 

What followed was billed a debate featuring Martin Oxley, Rebecca Stamp and Chris Branford – actually two short, snappy presentations and not an ounce of debate between them.

Martin Oxley made a plea for using the web for what the web is good for – emotional engagement and market research that matched this, and didn’t ask dam fool questions based on features and rational drivers, most of which were undifferentiated. He argued it is time that research took a few more risks and was constructed in less conventional ways.

Rebecca and Chris addressed the online panel and blog of Associated Newspapers, which has been used as part of the relaunch of the Mail on Sunday. The hybrid of panel and blog was good to hear. Blogs regularly form part of qual studies, but this collective blog sounded quite different. As the blog developed, participants could develop their own topics and agenda.

Perhaps the most tantalising development was the number of third party clients to whom they had sold access for posting topics to the blog. What? A media owner using research as an income generator from third parties? Way to go!

From psychographics to mashups

The last session before lunch was led by Peter Cape, with a plea for psychographic sampling. Again, if the web is about emotional engagement then why build samples with conventional demographics? This was a real problem – psychographic mapping showed how skewed the regular panels were. In a savage sideswipe, Cape attacked the blending of panels when there were major differences between them which would be likely to throw up odd and unpredictable findings.

Blending seemed too gentle a word here – this wasn’t a 15-year-old malt under discussion; mashup seemed more appropriate. So are psychographics the way forward? Alas no – because here a big fly plopped into the ointment.  Survey Sampling ain’t Linux, nor is Peter Cape’s alter ego Linus Torvalds.  Psychographic profiling tools are even more proprietary than conventional tools, and so aren’t about to bind recalcitrant panels together. Rather, they just show rather irritatingly how far the panels are actually apart from each other. Mash on then!

Response rates and data collection

After lunch, we had three client case studies: Luciano Juliano of Homebase, Sara Wroth of Royal Mail and Laura Chaibi of Orange. This was a useful session because it showed how modestly and sensibly clients are using the web. When challenged whether they were planning to set up online communities, all three replied coyly that this was in the works. But the full bells and whistles of Research 2.0 were not on show. 

Sensible projects included conversion tracking and competitive monitoring by customers, and moving tracking studies online. This is genuinely progressive, and is just as much evidence for Web 2.0 even if it doesn’t come across as flashy as some of the methodologies the research houses were clearly keen to grandstand.

Laura Chaibi galvanised the conference by getting everyone on their feet (it was the post-lunch graveyard slot after all) and conducting an elimination survey. She is fed up with people wanting to run surveys off the Orange website where the conversion to survey is 0.1% - needing traffic of 100,000 to deliver anything like a usable survey.

If you can only get one respondent out of a thousand, sit down. Then one respondent in 100, one in 50, and so on. I sat down at the one in five point, which made me one of the lucky handful of researchers she was desperate to talk to and work with. I was looking forward to telling her that I enjoy such ratios by recruiting people offline for good ole’ fashioned face to face groups.  But it wasn’t to be. Having trailed her cape, she dashed out of the door before we could trip her over with our business cards.

Online communities

Frederick-Charles Petit, of Toluna, then gave a presentation about using communities for online data collection, covering the well-known weaknesses of online panels and the potential of Web 2.0. He recommended using surveys embedded in many different sites to deliver huge panels and high levels of participation.

We had been told throughout the morning that response levels of online panels were under threat. But here was a magic bullet which increased participation by 40% or more: tell respondents the answers; let respondents ask questions of one another.

That said, the reason why panels are problematic is that clients need answers about paint, and postage and phone bills, which aren’t particularly interesting topics of conversation. And these active panels are huge: the UK panel is equivalent to the population of Greater Manchester!

New and old techniques

After the tea break, Ville Osterlund took the floor with a modest case study so radical that I had to run through my notes to make sure I had got it right. His agency, Interquest, had run an online community on green behaviour for three weeks, which had participants reporting behavioural changes in recycling and commuting. The credibility of the results isn’t the issue here: it is using behavioural change as a reported outlook from a research product.

The conventions of research mean we consider the attitudes and behaviour of respondents while we are studying them. But making long term behavioural change the point of the study is something else.  What if respondents knew that by participating in research that they might change? It might even increase participation if the subject matter was interesting or important enough. This seemed a startling example of Web 2.0. Research you might find interesting enough to volunteer for.

Nick Buckley followed with a provocative paper entitled “Why Bother with Online Communities?” which covered the technical and social aspects of communities. He pointed out that using old technology such as bulletin boards and chatrooms mean it is quite possible to get acceptable results – there was a continuum of methodologies of which creating real social communities was one extreme which was not strictly necessary.

International projects and trends

AJ Johnson, of Ipsos, finished the day with a practical session on running international online projects. His advice was grounded and practical, but having been warned of the dangers of blending (aka mashing!), the audience were a little rattled about the prospects of seamlessly aggregating panels across cultures and continents.

Mike Cooke rounded off day one with an audience Q&A, the warning that US and European dominance was under threat, and that new methodologies were far more likely to emerge in the theatres of China and India. Thus warned, we wound out to end of play drinks.

It was a good day overall: plenty of variety, albeit, yes, a little weary of the constant references to online communities, and niggled that a sprinkle of Web 2.0 pixie dust isn’t going to fix the problems of online panels, but there was a lot of pleasing evidence that online research is capable of producing fresh ideas, and surprising ones at that.

Continuing WARC's series of guest posts from speakers at Online Research 2008, Frederic-Charles Petit, Chief Executive of online research panel and survey technology providers Toluna, discusses how financial incentives are not necessarily the most effective means of securing the best results in the Web 2.0 world.

The market research industry is currently seeing a downturn in response rates to surveys completed online due to a dramatic increase in the number of surveys being conducted. A typical panel company’s reaction to this problem is to either raise the financial rewards for completing a survey or to increase the frequency of contact with the panelist via email survey invitations.

We believe that this approach is at odds with the best practices of online panel management; increases in financial rewards can only lead to the attraction of professional survey takers and the encouragement of multi-panel membership. Increasing the number of email invitations to panelists, in our opinion, can only serve to decrease the probability of response and data reliability.

To compound this global industry problem, the exponential growth in the number of Web 2.0-based websites has empowered the online consumer to be an active participant on the web, creating and publishing content at will. Consumers no longer passively digest information, but are now empowered to react on the actions of a brand. The online audience’s sophistication and technical competence has matured considerately since the founding days of the online market research industry.

As part of an investigation into the increasing levels of consumer web sophistication, Toluna conducted a research study in September 2007 of over 3,600 European panelists. The results clearly showed that the online consumer now expects to be enabled to express, collect, compare and share their opinions online. In fact, the results showed individuals are more interested in the ability to express and share opinions and gain social interaction benefits online, then perhaps more traditional Web 2.0 technologies such as video publishing.

A quick glance at any online panel will reveal a consistent trend in the panel industry’s approach to consumer and/or panelist communications. The vast majority of online panels base their brand communications and value propositions upon financial rewards. Further, such panels do not provide any technological innovations which would enable the free expression of opinion or member to member community interaction.

Effectively, such panels are creating membership, but in no way are they engendering a community or meeting the expectations of today’s matured web audience, even though such panels communicate with literally millions of individuals.

We have sought to increase panelist engagement and raise member responsiveness and data reliability by enabling our members to collect, share and compare theirs and others opinions. Members of our panel can create polls and post these to a 400,000 strong community and see demographic results displayed live in their browsers. They can also create qualitative debates on literally any topic for the community to respond to. Every poll created or answered is recorded as part of each panelists ‘organic’ profile history.

Each member has a personal community profile page, where their picture and polls and debate history is listed. Each panelist can link to a specific poll or debate created by another member, and add their voice to the ever-growing and contextually linked volume of panelist created research.

The website is now generating 120,000 poll responses per day and literally tens of thousands of qualitative opinions, all as a result of non-monetary based rewards.

With mobile, online and offline data collection all at marketers' disposal, the focus needs to shift from format-specific concerns to the big picture: how do you get the right information from the right person at the right time? David Day and Ralph Risk, of Lightspeed Research, address how SMS and mobile and online and offline research need to interact, and the subsequent changes to the traditional role of the market researcher.

The modern media mix and research

We sympathise with today’s marketers. Not only do they have more and more communications channels to reach consumers, but everyone seems to be busier and harder to find, more demanding and more marketing literate. In the past the researcher was king, using their powers to unlock the mysterious mind of the consumer. Today that is simply no longer true – the consumer is definitely in control and the researcher has to work harder and harder to understand them.

All these factors pose interesting questions for marketers planning to research customers, consumers and their target audiences. Nowadays, there’s a wide array of methodologies and channels to understand their customers. Traditional research methodologies are being rejected in favour of the latest trend – from using web chat forums to social networks, online panels to mobile. But how do you choose?

The answer is surprisingly simple; as with all good marketing campaigns, choose your methodology for your audience and your campaign objectives, your budget and your timescale.

To make these decisions, it is important to understand the limitations and benefits of each methodology, from depth of responses, interactivity, speed and of course, cost. Having a good knowledge will help you make an informed decision of what is best for your research and not just because it looks good. You may of course choose to use a number of methodologies simultaneously – this is a very useful technique for sense checking your answers.

Once you have chosen the methodology, it is important to design the survey to optimise the methodology. Don’t simply write one survey and use it across different methodologies and audiences: if you’re surveying under-16s, you need specialist help to write your survey; if you’re using mobile phones as the platform for research, you need a few short questions; if you’re using an online panel, make sure you’ve selected the right sample. And don’t forget to run a trial if you’re not sure that a new approach is right for you.

Continuing WARC's series of guest posts from speakers at Online Research 2008, new media pioneer and BBC columnist Bill Thompson, who is responsible for giving the opening address on the first day of the event, considers the complexity facing online researchers in a Web 2.0 world.

Web 2.0 tools turn a one-way medium into a two-way conversation, making the web far more dynamic and interactive. They make content creation simple and unproblematic and facilitate publication and the creation of connections between material.

At the same time, people are using the network and its nodes for many more aspects of their lives. The boundaries between offline and online, the real and the virtual, are becoming less meaningful, and just as we no longer really ask or notice if someone was in the same room as the person they were talking to or speaking on the telephone, soon we won’t bother to distinguish between forms of interaction or whether they are on or offline.

Integration between sites and services makes it a lot harder to figure out exactly ‘where’ someone is when they are online. I can be your friend at school, ADD you on MySpace, embed your videos in my blog and comment on your Flickr photos, trackback your blog posts and even message you from my mobile when you’re in your virtual world.

It also allows a rich online life that may be a delight for the people involved but will cause major problems for researchers.

I use 30boxes to put my calendar online and have it automatically update both iCal on my MacBook and Outlook on my desktop Windows PC. I post photos from my Flickr site directly to my blog. I Twitter from my mobile phone and see my Facebook status updated or make a mobile phone call from inside Second Life. A single conversation may begin with a tweet, move into Facebook and then email and perhaps end in a Seesmic posting or a real world meeting.

Anyone trying to understand online behaviour is going to have to integrate these various elements. If you want to understand who I am or what I am likely to be interested or do, or have any real insight into my attitudes, behaviour or likely purchases then you will need to understand all of these elements, and follow me as I sign up for new sites, services and tools.

Unfortunately the rate of technological innovation and the increasing complexity of the sophistication of the tools and the connections between then seems likely to ensure that research will lag significantly behind what is possible online.

As part of a series of guest posts from speakers at Online Research 2008, Graeme Lawrence, Business Development Director of Virtual Surveys, provides a brief preview of his presentation with Jo Carter, of United Biscuits, on their jointly-run online research community, and argues that researchers need to treat their respondents better.

These are very interesting times for online research, with much talk about Research 2.0 (or the industry’s response to the changes occurring on the internet). The core component of our Research 2.0 offering has been the development of our Online Research Community offer.

Snackrs is the community we run with United Biscuits, and I’m really looking forward to presenting with Jo Carter from the company at the conference. Our case study paper will highlight the following: 

• The need to stop treating the people who we research so badly. We need to avoid the term respondents, and start thinking of them as stakeholders, participants, and customers. We can’t keep abusing them with boring, repetitive research, and thinking we can keep the results secret.
• We have to work with all the stakeholders, clients, customers, researchers, other professions, and employees, in a collaborative way to create more relevant insight.
• The research from communities is deep, in the way qual can be deep, but is also based on the sorts of numbers that quant has always needed, opening new doors to insight.
• These techniques have major implications for clients, in particular the need to engage directly with the community members.
• Customers own brands and therefore need to be a stakeholder in the insight and exert greater influence in the brand relationship – clients need to cede control to customers.

As part of a series of guest posts from speakers at Online Research 2008, Tacis Gavoyannis, Strategy Director of Q Research and a board member of SwapitShop, looks at how new technology – and the further development of online and mobile phones – could change how market research is conducted, and lead to the development of 'a mass market of one'.

As part of a series of guest posts from speakers at Online Research 2008, Martin Oxley, of BuzzBack Market Research, discusses how the market research industry can move beyond the simple time and cost efficiencies offered by moving research online and truly engage with consumers in the Web 2.0 world.

If one examines the evolution of market research, it is apparent that there have been significant improvements in most areas. Until the switch to online data collection, every progression made in the predominant data collection method in the industry was expected to deliver – and received – more than simple gains in time and cost efficiency: significant advances were also made in the depth and breadth of insight generated.

But in the age of the internet, it seems that the same demands have not been placed on web-based market research. Despite having the most powerful, ubiquitous technological advance at its disposal, the MR industry has used the web basically to conduct research online as it would have done offline. So far, so bad: it seems that the only significant challenge made of the web is to deliver research faster and cheaper.

Given the amount of consumer-generated data that is freeflowing on the web already – and the growing number of consumers online, with their time spent on the web increasing – this opportunity is being missed by the MR industry at its peril.

Other industries such as online trading (which did not even exist 20 years ago) are engaging seriously with this opportunity. The global conversation is happening, as authors Locke et al. predicted in the Cluetrain Manifesto in 2000: 

A powerful global conversation has begun. Through the Internet, people are discovering and inventing new ways to share relevant knowledge with blinding speed. As a direct result, markets are getting smarter – and getting smarter faster than most companies.  

If the internet’s power is harnessed properly, there is little limit to what can be achieved. The MR industry has failed so far to engage consumers adequately online in a meaningful way. We are not asking for a revolution, just the application in MR of the intuitive technology that consumers already use to manage aspects of their life online.

Stuck in the 1990s?

The internet is evolving away from the Web 1.0/Introduction stage to the Web 2.0 phase of online activity and engagement. As a result, consumers are living in a world that is more technologically-advanced than today’s research is asking of them. These are captivated, engaged consumers who are self-motivated to share personal information via multiple web channels and enabling technologies.

The average visit on a user-generated content site is about 30 minutes, as long as the average TV programme. Through media like YouTube and Flickr, consumers are having brand discussions with one another, and cyberspace is replete with the voluntary data they’ve linked across their favourite sites.

But the market research industry isn’t even playing a cameo role in the film of this conversation, because it hasn’t been fast enough to accept that consumers are now in control of the conversation. (It’s a cliché, but the lean-back world of old is becoming lean-forward.)

Despite everything that’s readily available in terms of sophisticated web applications and tools, interactivity, entertainment and media capability, the MR industry hasn’t quite figured out how to use these to reach and retain the attention of the exceedingly cynical, distracted, research-wary consumer. In addition, no longer are industry players solely competing with one another for consumer focus and data exchange.

When consumers manage their data communication and consumption, even the best online research efforts are at a new continual risk: at any moment, responders are just one click away from thousands of other, more interesting websites. Let’s face it, in the research world, the only really limited resource is attention. Are you listening?

In the first of a series of guest posts from speakers at Online Research 2008, Mike Cooke, Global Director: Online Development, GfK NOP, shares his thoughts on the challenges and opportunities facing online research. Mike is the Conference Chairman, and will be speaking further on the issues facing the industry in his opening and closing remarks on both days of the event.

We are entering the second phase of the evolution of online research. The first wave was all about representation and the growth of online panels. It was about the migration of market research from the offline to the online world. It was driven by a collapse in the response rates achieved in the offline modalities as much as by the key advantages of online research namely cost and speed.

Online now accounts for almost one-third of survey research, and its share is still growing. In Australia, the USA and Japan is exceeds 40% of all survey research. Interestingly, here in the UK migration has been slower, with 21% of survey research now online. So we can expect a healthy immediate future for UK online researchers in the immediate future.

The current buzz is around “proprietary panels” – or “client panels” – which are growing at almost 40% year on year in the USA and starting to find traction elsewhere. This is because these panels sidestep some of the issues that are troubling access panel companies at present. These include recruitment issues, maintenance and coverage and data validity.

This has led to increased calls for quality standards and a sharing of experimental research. We will hear much about “data quality” at the conference I am sure, but it is the natural order of things for researchers to be querying these areas. That is how social science develops and the business of market research grows.

We are seeing the emergence of interesting new business models, including “real time sampling”, “panel exchanges” and “panel blending”, to meet these challenges. All of these will be discussed in the Conference and offer interesting access possibilities for the innovative researcher.

At GfK NOP, we have been using social networks and virtual worlds to conduct interviews and from which to build new research communities. We are entering a creative age in which the empirical rules we are familiar with will be questioned because the offline world will be challenged as much as the new to prove it is fit for purpose. Our “respondents” must become participants, co-creators of value or else researchers will fail our clients in their quests for consumer insights.

The second phase of the development of online research will be about engagement as much as access. Web 2.0 has given us a research platform to engage participants in the research process; it has democratised the internet and is allowing individualism to be expressed as never before in this new era of mass creativity and sharing.

Social media is a phenomenon, and its research applications will be demonstrated in the Conference. Researchers need to engage with social media, accept open source thinking and develop research “communities”, with all that implies for control.