I’m not a market research conference junkie (unlike Larry Gold or Bob Lederer), but I’ve been to a few. They all involve a series of trade-offs in the programme, between methodology, findings, the role of research, how to communicate it to top management, how to organize it, and so on.

This one had a tendency to be more about marketing than research, though the break-out sessions redressed the balance. This is not necessarily a bad thing, but it’s always worth remembering that the business of the ARF is research, and specifically advertising research – though that, as Sergio Zyman pointed out, is everything these days.

But what did it all mean? There were several pervading themes.

Digital confusion

The marketing and research – and, indeed business – worlds are trying to get their heads around the rapid changes wrought by digital technology, especially the web. There seem to be two responses possible: we can either sit paralysed like small furry animals in the headlights of the oncoming juggernaut until it crushes us; or we can run like heck and find a safe niche within it. Like well-trained managers of the old school, we’re too busy trying to control it. Let’s face it, we can’t.

Warp speed

The pace of change is fast, and possibly accelerating – though if that goes on too long we’ll be catching up with ourselves from behind. I prefer the old marketing device of re-inventing the wheel…. Seriously though, it is not possible to do some things at the speeds we are sometimes asked to, without massive compromise. Two strategies seem possible: either, we take, like Carla Hendra’s Ogilvy, the 70% solution, and correct on the fly – hoping we’ve got the direction, at least broadly right. Or we take the (reasonable) view that everyone else is going to botch the thing because they’ve not enough time to do it, and take a little longer to clobber them with a better mousetrap.

Silos
Everyone is busy demolishing or merging or consolidating their silos. All organizations are trying to become flatter, rounder, more 360°…. Yes, but isn’t there actually a simpler solution? It’s called teamwork, and management books have been on about it for years. Sure, the silo attitude needs to be eliminated, but specializations are still needed. Ogilvy might have it right.

Networks

People are networked. And networking. They always have: it’s just that the internet makes the process more visible. What does matter is to understand the nature of networks and influences within them, which is why the work of people like Duncan Watts and Mark Earls, and indeed their ‘adversaries’, like the good folks at Keller Fay and Malcolm Gladwell, matters, if we are to understand how to both research and communicate better.

Virtual worlds

Online research is virtual reality – of a sort. But not a lot of it, except at the experimental edges, looks very much like it. This seems to be missing a trick, or several. For example, given the predominance of the visual in the way we perceive the world, why is even online research still obsessed with verbal presentations and the dreaded semantic scale? (Of course, it’s because that’s what research has ‘always’ done, and what it understands.) The new stuff needs expensive and time-consuming validation.

Simplicity

We heard the mantra ‘keep it simple’ numerous times. Yes, that’s fine, but over-simplifying the complex is a mistaken approach. Let’s let the complex have its full value and stop chasing the single number solution (did anyone say Net promoter?).

Burrowing in the brain

In research technology, one of the next big things is the brain scanner. Well, maybe, but not yet. At present, it’s over-hyped, and results are over-claimed. There are indications that we may get somewhere useful with it over time, but it will need to become less costly and cumbersome if it is ever to go beyond a niche market area.

Quality

Quality is, rightly, a major issue for researchers and research users. We do need to be able to trust the findings that we get, especially where large quant surveys are concerned, and it’s good to see the industry re-focusing on this. Procurement departments please note that peanuts buy monkeys….

Engagement

Engagement has become the industry’s buzzword, even though the attempt to turn it into some sort of media metric seems to have fizzled out – arguably rightly. The trouble now is that it turns up everywhere, meaning different things to different people, in different contexts. Is it the same as, deeper than, or less than attention? Can you have engagement without attention? What’s involvement? How do they relate to relevance? And so on.

I guess this will be the last ARF Convention that I blog about, and this is my last re-think about one. Thank you for your attention (or engagement, or just for listening).

Quotes of the Day

Betsy Leichliter, Leichliter Associates:

"When we aadsked 15 year olds whether they had used Second Life, very few had. They found the concept of avatars kind of creepy. The real world was sufficiently compelling for them"

Stephen Kim, Microsoft's senior director of global marketing, on a successful response to putting real outdoor ads in a video game representation of Times Square:

"We literally had people coming to us, asking us to put more advertising in our game"

Colleen Fahey-Rush, EVP MTV Networks, on youth consumers:

"Theirs is a reward/award culture. Teens have grown up with the idea that everyone really can have prizes"

Day Three

If you're reading this in the late morning, it is possible this is your "info-snack", "dip into the digital candy bowl" or "virtual smoke break".

All three are terms Microsoft uses for the moment when, having completed the first major work task of the day, office workers reward themselves with a quick personal web surf, email, instant message or check on a profile page.

After two days and almost 20 speaker events, some ARF delegates might have seen Wednesday's morning general session as a similar reward.

Compared to the more technical and niche workshops at Re:Think 08, Wednesday began with presentations by Microsoft, Google and MTV into consumer trend research of the most accessible kind.

Afterwards, at least for this delegate, there was a return to three more unfamiliar worlds: the minds of 15-year olds, the successful use of social networking for brand building and the departure lounge at JFKAirport for the trip back home.

How many friends do you have?

One of the best demonstrations of the title of Wednesday's morning general session - "The Super-Connected Customer" - was also the simplest.

Stephen Kim, Microsoft's Senior Director Global Marketing, told delegates that in a survey of 14-24 year olds, respondents reported having an average of 53 "friends". Of these, they had typically never met 40 per cent in person. Instead, they connected with them through social networks, chat rooms, blogs and other digital phenomena.

A similar pattern was found across major international markets and reflected the popularity among youth audiences of using online social media to recruit and maintain friends/contacts/dates.

All three speakers agreed that this level of connectedness, facilitated by the ubiquity of mobile devices, instant messaging and social networking website, was a generational difference between teenagers/twentysomethings and their parents.

It may prove key to understanding how influence and brand preferences permeate through youth audiences, and to understanding audiences' expectations of how they want to interact with both media owners and marketers.

As an aside, however, it should be added that it is not clear whether this connectedness is a permanent change for the digitally savvy generation or a reflection of their current, relatively unburdened lifestage.

If domestic or career responsibilities increase, It is always possible these will eat into the time youth audiences currently have for maintaining networks of friends.

Also, as a later speaker asked, is the phenomenon explained by the fact that technology has enabled people to make friends more easily? Or is it that the type of people who have large numbers of friends are simply more likely to use social media to manage them?

Other digital trends

In his presentation, Mr Kim also argued that the pressure to juggle work and personal time was a major trend for most consumers.

From this Microsoft had embraced the idea that time-pressed people might see personal web surfing at work as a kind of "digital candy bowl" in the morning.  By lunchtime, consumers were more likely to be in an "escapist' mood - for example, planning for holidays or even retirement.

Mr Kim said this trend of shifting mindsets was true across media, not just digital, and marketers needed to reshape their communications accordingly.

Mobile, for instance, could cater for these changing moods and needs, and Mr Kim cited outdoor sites in Japan which used QR product codes (similar to barcodes) printed on posters to encourage users to use the codes to access brand websites on their phones.

In his talk, Jeff Levick, Google's director of global industry development and marketing, argued that the rise of digital media meant even US-focused companies had to think about their relationships with overseas consumers. This was because via the internet, foreign youngsters could influence their peers in the US as to their brand and product choices.

Although there were similarities across youth markets in different countries, brands also had to exploit local particularities.

As part of his case, Mr Levick compared dating sites in Turkey, China and India which required operators to adapt their services to local needs by, for instance, allowing Indians searching online for partners (and an estimated 48 per cent of online Indians visited dating sites last year) to filter results by caste and religion.

With 54 per cent of Indians under 20, and half of all Turks aged 28 or under, there was however plenty of room for such sites to expand, he said. Again, future growth could well come from the mobile internet rather than PC-based services.

In her presentation, Colleen Fahey-Rush, EVP of MTV Networks, ran through key findings from the broadcaster's research into youth audiences. Young consumers were analysed as:

• Masters of technology. The fact they can often operate devices their parents cannot, has shown them that influence and power can "flow from the bottom up".
  
  
• Expectant that media will be "a two-way conversation" which gives them access to content and freedom to air their views via postings, polls and other features.
  
  
• Often split down gender lines with boys working more as a group, and finding decision-making hard whilst girls adopted a more autocratic structure, typically allocating different roles to individual members.

Ms Fahey-Rush argued that MTV had tried to exploit these trends in a strategy entitled "The Power of Open" which gave viewers greater interaction and accessibility to MTV content.

In practice, this has meant:

• Adding features such as user-generated content, games and communities to its programming.
  
  
• Allowing users to embed MTV content into their web pages and blogs.
  
  
• Partnering with other websites as well as launching 100 new websites in 2007 alone.

She cited the record traffic to the website of the most recent MTV Video Music Awards show as evidence the plan was paying off.

Teen Trends

Similar themes were addressed by three further presentations.

The first, co-presented by Leichliter Associates and Abbott Research, was based on an 11-country study of how 15-16 year-olds use technology to communicate with each other.

It found:

• Similarities between groups in different countries outweighed differences.
   
• The mobile device was central and obtaining one was "almost a rite of passage" for teens.
  
  
• Teens chose between providers of instant messaging or social networking on practical grounds such as the value, user-friendliness and audience size of the service in question. They were prepared to jump between providers, if needed.
  
  
• Podcasts and blogs were less important than picture sharing and instant messaging.
  
  
• There was a focus on fun, open-ness, tolerance and often marked levels of courtesy.
  
  
• Phones were often the last products teens used at night and the first they did in the morning. In the afternoon, respondents could often be found scrolling through phone contacts or messages without texting or ringing anyone. The latter may be an untapped opportunity for marketers to exploit.

Data on a broader audience was provided by the next presentation. In this Harris Interactive outlined results from two separate studies into digital media usage, and specifically websites dealing with movies and cars.

It found that:

• Reach of different types of websites was highly variable among different audiences with only YouTube, Wikipedia and to some degree MySpace having what it might be called mainstream reach.
  
  
• It was difficult to generalise about audiences: eg. there was a substantial older audience for chat rooms.
  
  
• Perhaps a third of surveyed users didn't do anything online during the period surveyed.
  
  
• Technology was not eating into time used for physical social interaction, and was complementary to mainstream media.
  
  
• Technology was contributing to word of mouth, but not necessarily originating it and people still relied on trusted friends and contacts for recommendations.

C2C to Campaign

By comparison with traditional media, there are self-evidently fewer case studies for brand-building in online social media. The final sessions I attended provided some metrics and actual examples from this genre.

First, Keith Brady, of Marketing Evolution, a company which carried out a well known study ("Never-ending Friending") into branding on MySpace, outlined some measurement metrics used for campaigns on  online social media.

It defined online social media campaigns as having both a Business to Consumer element (B2C), and a Consumer to Consumer element (C2C).

Metrics for B2C included straightforward cost per impression figures for brand banners on social sites. For the C2C element, there was tracking of how many people downloaded of branded material and forwarded it to peers.

According to Mr Brady, an online social media campaign's ability to create brand value could be assessed by measuring similar metrics to traditional campaigns, such as awareness, positioning, positive sentiment, purchasing intent, loyalty and advocacy.

A case study of an Adidas campaign, run by CARAT on MySpace during the 2006 FIFA soccer World Cup, was then presented. The campaign allowed users to download and pass on a variety of branded material, submit original content for an "impossible goal" segment and participate in other ways. A total of 21.5m estimated impressions were created by the campaign.

Adidas ran a second campaign using a different theme which allowed participants to ally themselves with one of two fantasy football teams.

Post-campaign tracking, suggested that up to 70 per cent of the brand value created by the campaigns was through the content forwarding effect (known as the momentum effect) rather than the more conventional brand impressions measure.

Learnings were summarised as:

• To maximise consumer to consumer interaction, brands need to be seen as leaders in their field (Adidas and World Cup soccer).
  
  
• They need to give people a reason to talk about them by creating special content, widgets and other features to download and forward.
  
  
• They also need to offer participants the opportunity to realize fantasies and aspirations (similar  MySpace campaigns have offered music acts record deals etc).

The limitations of such campaigns included:

• The offline impact of social media campaigns could not yet be measured.
  
  
• Making interaction simple and templated was seen as key, but could limit flexibility.
  
  
• Not every brand is right for social media because it may not be seen as sufficiently engaging or as a category leader.

This last point emerged as an overall theme from many sessions at Re:think 2008.

Many of the more prominent examples of digital investment, and to some degree digital success, featured groups such as ESPN, Adidas and MTV which are well-funded, category leading, and possessed of access to engaging content around which to build digital campaigns and services.

Whether smaller or low-interest companies will be able to make the same success of digital is perhaps the question for future conferences.

Connected consumers

Mary Ann Packo of Millward Brown moderated presentations by three experts in new media and how they connect the world’s consumers.

1. The Microsoft view

First up, Stephen Kim (Microsoft, and the new ARF chairman) used research across 14 countries to illustrate the changing habits of an increasingly media-savvy public – especially the young. The bad news for advertisers is an increasing aversion to ads; the good news is that they are positive about good ads, and willing to share them around.

Much of what Kim talked about came from essentially ethnographic research. He focused on how the real world maps onto the online world: on the one hand, we seek to create an online neighbourhood, where things are familiar and safe; but we also use online to explore, which is a different mode.

This applies, equally, to how we manage our increasingly jumbled work-life relationship. When we arrive at work, the first thing we do, usually, is take a little time to sort personal affairs: then we work; then we get to what Kim called the ‘virtual smoke break’ or the ‘digital candy bar’. These various modes suggest different types of ad opportunity, for different products, geared to time of day and situation.

Relationships, too, have become digitalized. Research shows that the young, globally, claim 53 friends, on average.  Hardly the picture suggested by Holden Caulfield in Catcher in the Rye. But closer study shows that they have a more realistic  6 close friends; 27 not so close; and an unexpectedly large 20 online friends – people they have never met and speak to only digitally. In some countries, like Brazil and India, the online figures are even higher.

2. The Google view

Jeff Levick (Google) took a deliberately international perspective, looking at a world with 1.4 billion internet users (just 5.2 billion to go…); people whose search interests focus on other people, but with different biases. If advertising means connecting brands to people, we have to understand the people, locally – not just globally – and recognize that as technology has created new connections, everyone both produces and distributes content – including advertising.

He illustrated cultural difference by looking at dating in India, Turkey and China. In India, no less than 48% of male internet users (a minority, in a young market) had used the net to search for a marriage partner. In India, social networks combine friendship and dating. In Turkey, an emerging market where internet usage is still quite low, the leading sites are all in English – brands which have almost accidentally become global (YouTube, Facebook). In China, where a small user group is deeply engaged, blogs are a key medium for social interaction, with lots of online dating in a market where by 2020 there will be 30-40 million more marriageable men than women.

Jeff stressed that the possibility of becoming global by accident is always there, and that brands need to be ready to adapt. And the information to help this is there – ‘guessing is so 90s’.

3. The MTV view

Colleen Fahey-Rush (MTV) gave us a kaleidoscopic view of MTV’s cross-platform multi-channel offer, constantly-changing in response to research which they are happy to share with advertisers. Young consumers are these days ‘techno-creative media masters’, who feel both entitled and empowered by digital media, and this has led to a democratization of influence – a process that works rather differently between boys and girls.

For MTV,this has led to strategy recognizing the ‘power of open’ – media seen as conversation, which means listening as well as talking. Out of this, MTV has recognized a strategic need to be everywhere their audience is; to allow consumers opportunities to create, from UGC to virtual worlds, via interaction and mobility; and to build vertical entertainment properties that work across a variety of platforms.

In the subsequent discussion, several themes emerged:

• The need to recognize and relate to consumers’ fluctuating mindsets, and fit communications to both medium and context.
  
• Hence, to move away from monolithic ad campaigns and single-media plans – to make media analysis far more granular.
  
• Change as a constant: if you haven’t changed, and are not changing, you are out of it
  
• Relevance as central to successful communication: and this means fuelling consumers’ conversations between themselves, as much as any dialogue between brand and consumer.
  
• O n a global scale, get down and dirty: understand the local environment, and adapt to it
  
• Nonetheless, consumers haven’t changed (much), but the media platforms have.

Contemporary media habits

Following this, I went to a forum on contemporary consumers and their media habits, introduced by Amy Raihill of Shell, who admitted to struggling with consumer and media fragmentation. The first paper came from Judith Ricker and Scott Evans of Harris Interactive, who looked at consumers’ likely acceptance of mobile advertising, based on Harris Polls done online in 2007 and 2008.

As  they pointed out, the US mobile market is undeveloped in comparison with Europe’s, let alone the trailblazers of east Asia and the Nordic markets. But structural changes mean that it may be catching up. What is very clear is both resistance to advertising on a most personal medium, with 64% of adults and 45% of teens saying they would not accept ads under any circumstances.(So much, you might say for the much-touted ‘next big thing’.)

What’s more, even those who would accept ads want to remain firmly in control, with the ability to opt out; with rewards for being in; with the ability to define what ads they will accept. In a relatively low-tech environment, there was a clear preference for text messages, and expectations of practical ads – most wanted were offers on ding by adults, and entertainment possibilities by teens.

But new business models and new features may change all this, so watch this space..

Next, Howard Horowitz and Adriana Waterston of Horowitz Associates gave us an overview of the urban TV and technology consumer. ‘Urban’ covers 68% of the US population and includes suburbs as well as cities, but urban consumers are disproportionately multi-cultural, and account for over 80% of Hispanics, African Americans and Asians. A detailed paper with masses of figures followed, broadly showing that the minority groups (who actually account for over 60% of the urban population, in total), have much the same media/tech habits as urbanites as a whole, though there’s a bit of a ‘digital deficit’ among blacks.

Perhaps the key point to emerge from all this is that – as ever – content is king, and the minority groups tend to be under-served by material that is language or culture-specific and relevant. The coming digital transition should be empowering in this respect, and present opportunities for marketers to reach the ethnic ‘long tail’. A ‘virtuous circle’ of technical access is driving consumer demand, which focuses on entertainment.

Finally, Nancy Bates and Mary Mulry of the US Census Bureau treated us to a paper on the census, and how the Bureau works to ensure that everyone is reached and counted, cost-effectively – every 1% increase informs mailed back (as opposed to having to be pursued by mail or personal follow-up) saves $75m. for a service on which government and business depend.

They described a professional but broadly straightforward process by which the population was segmented at the Census tract level to enable a calculation of ROI from the imputed increase in mail returns at the 2000 census. This enabled the Bureau to plan with their agency how best to allocate effort, and the application of the segmentation to industry media research guided media/contact choice.

Word of what?

After lunch, four forums looked at current ARF research initiatives. In the word of mouth forum, Mark Earls (Herd Consulting) and Scott Thomson (Naked) looked at the issue of how important WOM actually is. Clearly the idea that people are connected and this contributes to marketplace and other effects is true – but how much is due to WOM, and how much to imitation, much of which may be unconscious. Drawing on behavioural science, Mark argued for the importance of copying as a source of all kinds of learned behaviour, and for the proposition that most marketing initiatives fail because they are using the wrong ‘map’.

He also questioned the concept of the ‘influential’ as marketers and researchers usually try to apply it, arguing that there are informational ‘hub’ people, but that they don’t in fact spread behaviour. While not dismissing WOM, he felt that the balance was wrong, and imitation should be given more importance. (This seems to square, somehow, with the fact that most of the information we take in is visual, rather than verbal). In sum, we need to re-purpose research, with better ‘maps, and to recognize that the maps are dynamic – for example, segments aren’t fixed.

Scott took a more practical view of the questions raised, illustrating his comments with Romanian President Ceausescu’s final speech, as the coup that felled him gathered in front of him(a failure to recognize the ‘social geology’), and trend charts that showed the popularity of the democratic candidates – including the recent Clinton ‘dead cat bounce’. His point was that we should track events to identify moments of traction or receptivity, and map social geology. From the uncertainties inherent in this, we should develop options, not grand plans, and be prepared to take advantage as consumers reframe value.

Following this, Duncan Watts (Yahoo!) proposed an experimental programme to investigate the importance of influentials. As he pointed out, there are at least six distinct ways of identifying influencers, and some or all of these may be right. What is needed is a properly formulated real-life experiment to establish whether strategies based on influentials really are better than random or none. All that’s needed are volunteers and money. All offers gratefully received by the ARF.

End-point: the store

Joel Rubinson, new Research head of the ARF, introduced Roman Portilla from WalMart to moderate a session on shopper marketing, pointing out the significance of the store as a communication point for marketers. Portilla repeated that you can’t manage if you can’t measure success, and said that WalMart had found that while they could measure a great deal, they really did not have a handle on the basics of store traffic, or on how they could really delight customers throughout the store. This had led to their enthusiastic support for the PRISM project, which was the focus of the following presentations, by Peter Hoyt (Hoyt Publishing), George Wishart (Nielsen) and Kelly Downey (Unilever).

Hoyt led off by outlining the changes in the consumer and retail markets that had led through brand management to category management and now shopper marketing. This had focused priorities on the in-store environment, which had to justify its role as a medium alongside other media. CPG companies were not geared to take advantage of shopper marketing, nor were many retailers. Hoyt’s In-store Marketing Institute had mounted a pilot to measure in-store, and this had led to the larger PRISM project.

Wishart took up the tale by showing the PRISM sponsors – manufacturers, retailers and agencies, who had combined primarily to establish common metrics to measure consumer reach in-store. This involves measuring the entire store, with sensors in the aisles, and auditors to count traffic in places the sensors could not cover. The data can be combined with scan data and household panel data to provide: gross impressions by location, by demographic, etc; details of promotions in place; and a variety of specific applications for retailers and advertisers. Following a large-scale test, the project is due to roll-out nationally.

Finally, Downey gave us an indication of how Unilever expects to be able to exploit PRISM findings, to improve targeting and positioning of promotions, to identify particular stores to proritise, and so on. At present she said they are still at the ‘crawling’ stage, but the potential to contribute effectively to the business, and to the relationship between marketing and category management is evident.

And the online shelf-space?

The final keynote came from Calvin Schmidt of Johnson and Johnson, who gave us a densely-charted rundown on why online presented a retail opportunity and the possibilities open to advertisers – not least to disintermediate. Quoting ‘People don’t buy brands, they join them’, he took us through a range of material and prescriptions, which he summed up in five key points:

1. Start with the consumer – what does she want?

2. Don’t follow the leader – leverage your brand’s benefits

3. Brand value is ultimately determined by the number, type and quality of interactions and transactions between consumers and your brand

4. Leverage developing technology – but do it strategically, not by responding to the latest tempting gismo

5. Measure everything, learn and update.

Quotes of the Day:

David Calhoun, CEO The Nielsen Company:

“Every CEO in the world is just trying to see around the corner a little more clearly than the other guy. Seeing around the corner is what this industry should be about.”

Jim Nyce, SVP, Consumer Insight and Strategy, Kraft Foods:

“Collectively, we took our eyes off the ball with relation to internet quality data”

Presenter of Brand Intel session:

“There are drugs which warn of their potentially lethal side effects that gained better scores than the PlayStation 3 did with some of our respondents”

Kanishka Agarwal, VP New Products, Nielsen Mobile:

“There are 14m Americans who pay for access to mobile video, but only 4m use it. I can’t understand it: mobile video is so expensive. Maybe they are all charging it to their company.”

Day two

The second day of Re:Think 2008 provided answers on the potential for marketers of online social media. But these are unlikely to be welcomed by the more excitable followers of the social media genre.

There was also an industry sit-down of the chiefs of the major market research groups and discussions on the importance of integrating audience data across media platforms which were, ironically, at separate ends of the timetable.

Caught by the web: social media as a research tool

Two presentations challenged accepted definitions of social media. The first, by Brand Intel, was the completion of a two-part presentation into online social media as a potential source for conducting product and brand research.

Part one set out the company’s research methodology. Broadly, this is based on searching unsolicited consumer postings on online social media for discussions of particular keywords related to the subject studied.

The company defines social media as special interest websites with multiple discussion groups such as Edmunds.com, the automotive website, or the gamer-targeted Gamespot.com, rather than blogs or FaceBook. It argues that the specialist sites are more likely to contain message threads on brands and products rather than the solipsistic musings of bloggers.

Keyword-related search results are filtered and matched for relevance. The results are then coded by researchers as to the product, attribute and sentiment mentioned.

Brand Intel cited three hard-to-reach research audiences where this online social method could improve on mainstream techniques:

• Sufferers of Crohn’s disease: Because of the low incidence of Crohn’s disease, Brand Intel maintains it would be costly and time-consuming to recruit enough respondents with the condition using only conventional research channels. However, hundreds of websites – some segmented into new and long-standing sufferers, carriers and relatives of sufferers – discuss the condition, allowing the company to research the subject relatively quickly and cheaply.
  
  
• Gamers: video gamers were also difficult to research traditionally, it said. Often they were not at home to interviewers, were unwilling to complete surveys or were young enough to require questioners to gain parental consent first.
  
  
• Upmarket car buyers: with car buyers, it argued, that the number of annual buyers of some models was so small that recruiting a big enough sample would be costly and slow.

However, it cited the following limitations to online social media:

• Consumers effectively  chose the topic and the particular attributes under discussion
• Some products do not engage consumers sufficently to produce much discussion
• Market size is less important than the engaging nature of the product or subject under discussion
• Consumers can’t discuss what they don’t know
• With new products, commentary comes from early adopters who may not be typical  

In Part Two, the company provided case studies on Sony’s PlayStation 3 and Apple’s iPhone.

In its study of pre-launch discussion of the Sony console, it found that discussion of PSP3’s price accounted for a significant chunk of internet postings and was often negative. The opposite was true of Nintendo’s Wii. Although mainstream media and traditional surveys forecast Sony would win that particular round of the console wars, Nintendo emerged on top.

A similar study of pre-launch discussion of Apple’s iPhone found that price was a negative factor for the device but outweighed by positive reaction to its design and a strong intent to purchase among commentators.

Consumers were unsure about the device’s touch keyboard. However, post-launch research suggested this became less of a problem with usage. By most standards, the launch succeeded: Apple sold an estimated 4m phones in the first year.

Overall, Brand Intel claimed its method should be used by brands to compare their standing among consumers against that of their rivals, as well as to assess product and brand perceptions at different stages in a product’s life cycle.

Social media: a question of trust?

In another dip into the “social media sandbox”, Pollara Research laid out its view of the genre. Again, it defined online social media widely. It counted eBay, Amazon and discussion sites of newspapers and broadcasters as well as You Tube and My Space.

Its presentation was heavy on numbers, drawn from separate studies into views of online social media among business executives and consumers.

Pollara’s main thrust was that while awareness and claimed use/knowledge of online social media were high, the genre scored poorly as a source of trusted information. It said web users relied on recommendations from family, friends and peers but were suspicious of active, “noisy” commentators on such sites or bloggers as biased sources.

Visitors to social sites told Pollara they were predominately interested in keeping in touch with friends on such sites. They were open to finding brand and production information in social media, but were looking for a general current of information or a pointer to other websites rather than a single, strongly-held position such as a blog might provide.

The belief that marketers could tap such web media by identifying key bloggers or “influencers” on social sites and co-opting them into transmitting brand messages on their behalf was flawed. Robert Hutton, executive VP of Pollara Research, said: “The idea that there are a few strong influencers on social media is not borne out by the results we’re seeing.”

The studies also uncovered considerable levels of resistance to brand advertising on such sites, particularly among the 18-34 year old audience group. Almost half of some surveyed groups thought such sites were already too commercialized.

Where this leaves Facebook and others which have valuations based on the assumption they can sell their huge audiences to advertisers is anyone’s guess.

3. The challenges for mainstream research

By comparison with social media, the challenges of the market research industry seem relatively modest. A morning session which included the heads of the Nielsen Company, TNS, Kantar and GfK and senior clients from Washington Mutual, Google, Citi and Kraft Foods, achieved a fair degree of consensus on what the industry should prioritize:

•   Hiring more people with knowledge of client industries and better insight skills
• Integrating data collected from different sources
• Guaranteeing quality of data responses, particularly from the internet
• Helping marketer clients “sell” research to their company boards
• Retaining and developing key research talent
• Creating partnerships between clients and suppliers on the aims, costs and practice of research

So far, so uncontroversial. From the clients, however, there were complaints about the disconnect between what consumers told research surveys and their actual behaviour, and about occasions when clients had had to carry out work to rectify problems in results provided by research suppliers.

The session moderator – Stan Sthanunathan, Coca-Cola’s VP Marketing Strategy & Insights – probed away at the question of who should take responsibility to ensure the quality of survey responses.

Jim Nyce of Kraft Foods, reminded delegates that in the 1980s clients frequently took the lead in checking and validating responses from research surveys. He cited his former employer, Quaker Oats, which three decades ago had a 100-strong research department which called back 50 per cent of all survey interviewees to validate their responses.

Since then, outsourcing and the influence of procurement, he said, had left managers under pressure to produce research quickly and cheaply, and their suppliers pressed to deliver it to tight deadlines.

Eric Salama, CEO of Kantar, argued that clients still had the power to shape the debate by rewarding good work and punishing bad.

A questioner from the audience asked whether any of the research executives on stage had ever refused to carry out work because a client was not paying enough or the timetable was impossible.

Mr Salama cited a current situation with a telecoms company client which was demanding an impossible timetable leading to what he described as “incredibly tough conversations”.

David Calhoun, CEO of the Nielsen Company said it would ensure its clients knew that if they demanded an unrealistic timetable they would be likely to receive flawed results. If they accepted this, Nielsen would continue on with the work, he said.

Pressed as to what the company would do if it believed its work would be misused by clients, Mr. Calhoun said he believed “9 times out of 10”, the company would refuse to do the research.

Data integration: together at last

And so to data integration. Since several delegates pointed out that bodies such as the ARF have been discussing the potential to combine data from different media and product surveys for a decade or more, it seems unlikely that this year’s contributions will seal the issue.

The end of the “Project Apollo” collaboration between The Nielsen Company and Arbitron - which failed to attract sufficient advertiser support for its bid to provide a single source of US media data - has not dampened appetites to pursue data integration.

A morning session flagged up that the subject would be tackled by the ARF’s media council over the next year. For the uninitiated, like me, there was also a quick primer on the difference between data integration and data fusion.

Integration, it seems, is the process of combining information from two or more separate data to estimate accurately results not available in any single data source. Meanwhile, fusion attemptsto connect a given respondent in one survey to a respondent in the other survey whosecharacteristics match most closely on a pre-selected set of variables.

In the afternoon, there was a more hands-on approach to the integration issue when Nielsen updated delegates on changes in measuring audiences on home television, out of home TV, mobile and the internet.

The two-hour session took the form of Nielsen executives outlining recent initiatives on different platforms using the recent “March Madness” series of NCAA Basketball games as a test case for measurement:

• In the set-top box market, the company is working with cable and telco companies on installing its measurement equipment in installed boxes – now in 2m US homes (though only a small number provide data from the box via a  return path such as a telephone line)
  
  
• Set-top data allowed the company to provide some assessments on use of High Definition TV. In “March Madness”, for instance, an estimated 13 per cent of viewing was via HD
  
  
• In Online, Nielsen has been working with different techniques: panel measurement, measurement of streaming and measurement of online buzz etc
  
  
• During “March Madness”, measurement showed people still preferred to read HTML text reports on the web rather than view games online
  
  
• The Beijing Olympics may prove an inflection point push internet video usage into the mainstream because the 12 hour difference for US viewers means they may want to watch live events from China online from their office desktops
  
  
• In Mobile, video audiences are still small but Nielsen has been collecting data by methods, including examining mobile bills for download usage (with subscribers’ permission)
  
  
• Nielsen now provides data for audiences across TV and out of home which eliminates duplicated viewers, though mobile has not yet been added to the other two.

Online research quality?

Early birds were able to catch an update by Robert Tomei of TNS on the ARF Online Research Quality Council programme. This has assembled a powerhouse of researchers with a matching body of blue-chip, client-side equivalents to address four main issues:

• multi-panel participation;
• panellist engagement and interaction;
• the reliability of online research; and
• agency transparency and quality metrics.

To explore these issues, four working committees are looking at, respectively: the definition of quality; accountability for quality; transparency and quality metrics; and funding and outreach.

The key short-term deliverables, which are the subject of current activity, are research-on-research around quality issues, especially the question of multi-panel respondents and the development of a 'dashboard' to enable agencies to report panel performance and, ultimately, results.

Research CEO forum

Next, Stan Stanunathan of Coca Cola introduced a panel of research CEOs, consisting of David Calhoun (Nielsen) David Lowden (TNS), Eric Salama (Kantar) and Klaus Wubberhorst (GfK).

Stan's intro focused on the challenges of constant and accelerating change, where the MR industry itself had consolidated, spawned new boutiques, adopted the internet (which itself introduced new competitors), and a decline in quality control by clients, leading to issues of reliability in research design, data and analysis.

David Calhoun said that Nielsen continued to integrate, with most investment going into consolidating online investments and the heroic and difficult effort to eliminate boundaries between what had been 26 distinct businesses. Integration, however, had to go beyond this, into a willingness to become open and integrate client data. Open systems are hard to develop, and even harder to get in tune with his other aim - to keep things simple. He declared himself more optimistic than ever.

David Lowden stressed the importance of generating insights that could affect business decisions. And since most CEOs have engineering or finance backgrounds, research has to be translated into new language and new angles - which requires new investment in talent, experience and training. At the same time, data collection faces challenges of quality, both of samples and data.

Eric Salama talked about quality - of engagement with respondents, of talent and of insights. The gallery of faces that kept him awake at night each carried a lesson:

1. Stan Stanunathan: a demanding but fair client needing consistency and quality on a global scale: something that demanded better ways of engaging with respondents. Are we experimenting enough? Are we giving enough back for their data?

2. Stan's CEO: A man with many worries, for whom the answers need to be simple. Are we simple enough for our clients' internal client?

3. Roslayn Tann, CEO of RI, Singapore: Are we doing enough for her, in terms of training, mentoring, help and career development? Kantar's training budget had trebled - but was this enough?

4. Einstein: A quote from Kantar's creative director (a radical new post): 'The intuitive mind is a sacred gift; the rational mind a faithful servant…' Insights are there to be found, with the promise of the huge consumer power of the internet to help: how can we best exploit that power?

Klaus Wubberhorst wondered if anything had actually changed. MR's task remains much the same - but data collection and analysis methods were changed by technology. For clients, this looked cheaper, easier, faster. For agencies, it looks more complex, demanding and multi-choice. He worried about talent and training; how to leverage technology for greater efficiency; and the changing quality landscape, due to the internet.

All this could involve trade-offs between quality and price - fine if this is transparent: there's nothing wrong with cheap, quick and dirty research if everyone agrees that's the task.

The client-side perspective

The discussion was then joined by a clientside panel, of Jim Oliver (Google), Ron Galley (Washington Mutual), Jeff Todder (Citi) and Jim Nyce (Kraft Foods).

They individually stressed some specific issues:

• the universal desire for data-driven decisions, matched by a huge increase in data;
• the need to ensure good data and track data quality over time;
• the fact that the US's MR spend to adspend ratio puts it about 20th in the world - are we-under-researching?;
• are researchers being as customer-focused as the clients are trying to be;
• client responsibility for research quality has evaporated, while procurement has intervened - though it is becoming more sensitive to quality issues.

An extended discussion of research quality included the idea that perhaps there should be a ten cents per respondent fund to be hypothecated for research on research - and the worry that this might lead contributors into uncompetitive pricing. Problems of response and of the excessive share of online response attributed to a very small proportion of the online community were not seen as an entirely new problem. We have had focus group professionals for years - and I could add declining response rates to all kinds of research.

Key, perhaps, to the issue of non-response lay in David Calhoun's wish for research to be 'more like Facebook' - more imaginative question design (in 3D , colour and audio-visual terms).

Neuromarketing - getting brains in shape

Two research presentations I saw looked at using brain scanning to shed light on advertising - the embryo science of neuromarketing. Bill Knapp of FKF gave a paper on FMRi scanning and engagement, while Stephen Sands of Sands Research presented a hybrid FMRi/EEG technique.

Given two different approaches, it's no surprise that their definitions of engagement were rather different, and the focus of what was shown varied. Bill was more bullish about what can be achieved by putting a respondent in an uncomfortable box for an hour or so, with a temptingly simplistic map of how different parts of the brain respond in different mainly emotional ways.

His key points were two:

1. That what he calls engagement has to fight against counterforces - eg attraction against guilt or anxiety - and that FMRi can help find ways to reduce the counterforce.

2. That we actively filter the stimuli presented to us - so that it's important to check that this is not happening.

Stephen took the view that FMRi is both a slower and in some ways cruder tool than EEG, and that putting the two together, though expensive, could improve results. He recommended using the method for small-scale validation research.

On the evidence, this is still a very experimental area, and it is by no means clear that it can yet add much to what can be obtained by traditional research methods - in spite of claims to the contrary.

Shopper marketing

James Sorenson and Melinda Smith of TNS gave us a quick overview of the shopping environment and the different forms of consumer contact it offers - some 20+ in all. These could be measured through research in terms of which people see as most influential and which are best at differentiating brands. From this, it's possible to map contact points on an influence/differentiation matrix, and see how these might be exploited to better present a brand.

A lot of this involves trade-offs between cost and effectiveness, and they offered us some useful insights into display location in relation to traffic flow, and the unexpected view that in-store TV ( a strong performer) is better placed relatively low down, which is where shoppers' eyes are focused.

Clarifying the digital confusion

At lunch, David Kenny, Chairman & CEO of Publicis's Digitas division – introduced as a great remover of digital confusion - gave us a very brief show of statistics that confirmed that digital is large and growing. He said he expected advertising to catch up with this growth in the next 3 years, although it's very complicated territory. He showed us the Web Trends internet map, which looks a bit like the London Underground mated with the New York subway; and marketers want structures and metrics to manage their dealings with it.

Against this, his theme was that the web is very personal: the top 5 global sites are two search engines and three social networks, and the networks are providing users with an emotional access to the web.

From Digitas's viewpoint, all web communication has to be built on three pillars: create, distribute and optimize.

1. Create: His illustration here was the Holiday Inn NASCAR Go Green campaign for Intercontinental, where at least among the loyalty card members, interactivity and involvement with the website had become a hyper-active form of the members 'doing their own marketing'.

2. Distribute: here the illustration was Miller's caffeine/beer beverage for young drinkers, Sparks. With a target audience who rarely view TV, Sparks is tailor-made for social networkers keen to 'live the liquid life', and he showed a wild video of examples of orange-tongued partyers lifted from Flickr and Facebook, stimulated to try the brand by a mostly online campaign. Sparks drinkers had made the brand theirs.

3. Optimize: finally, he looked at how GM had used focus groups and interviews to generate vox pops about their trucks, and integrated this into online messaging, fuelled by all forms of word of mouth - 'listen and learn'. The monitoring of customer behaviour and comment enabled the campaign to tailor communications to the individual, mapping product offers to perceived interests.

All in all, David believed that the internet enabled people to build brands for themselves, and empowered them to do so. It is up to marketers to swallow their thoughts of command and control and go with this flow: the job is to help it to happen.

Making babies: the European mothers market

As a complete contrast, Hélène Lepetit and Patricia Gelin of L'Institut des Mamans told us about Europe as a market for mother and baby products. Why? Because it's very large, with 80m children under 14 years old and 5.2 m babies are born annually - some way more than the US figure of 4.3 m. Births are not merely a happy event - they stimulate buying: moving house, new cars, new appliances, cameras….

The market, however, is far from homogeneous, though it has the common facet of first births being to older women (average age 28+, against 25 in the US) and as women age, procreation gets harder, so babies are more precious.

There is a fairly clear North/South split, with the Northern mothers more modern in attitudes, more family-focused, less likely to be married, more likely to work, with higher child benefits from the state, readier access to childcare and to maternity leave, so birth has less impact on income and working .

If we look at the big four countries (Germany, France, UK, Italy) and Sweden, which is the parenting 'thinktank' of Europe, the media provide a remarkably varied picture of mother-baby relationships, as reflected in the abundant parenting press:

• France : motherhood is idealized, and a responsibility, to bring up the 'perfect child'. Standards of care are rigorous, as illustrated by a 'zero alcohol when pregnant' ad, and ads and editorial show an intense almost animal relationship between mother and baby, with a clear undertone of purity and the myth of 'original motherhood
  
  
• UK : babies here, by contrast, are treated as 'real people', with plenty of ethnic minority representation in ads and editorial, and a playful tone to the ads. A lot of material focuses on risk avoidance, but it is not unusual to see a pregnant woman holding a glass of wine - even though the text says 'put it down'.
  
  
• Germany : for Germans, a baby is part of a family unit, albeit idealized, and the child's education is a mission for mum.
  
  
• Italy : Italy sees a 'complete' woman, aspirational but not idealized, and working. She has a symbiotic and sensual relationship with her baby - lots of touching and eye contact, and is often shown at the centre of a large family, though Italy has one of the lowest fertility rates in Europe.
  
  
• Sweden : for the Swedes, parenting is given great support by the state, with generous maternity and paternity leave, and childcare from age 1. So, 80% of mothers are working, while the latest initiative is to encourage women to have babies earlier - while still at university. Swedish parenting magazines are segmented titles, with one title having separate issues for pregnancy, new born, first year, and so on.

To access this complex but valuable market, L'Institut offers a range of research products giving access to 400,000 mothers, with a quick turn-round online service in addition.

The word from Nielsen – integrating TV, PC and mobile research

As the final act of the day, a specials session was introduced by Paul Donato of Nielsen, outlining how Nielsen's policy of integration, is being put into practice to provide a coherent overview of consumer viewing across all three screens - TV, PC and mobile.

As became clear, this is a project in progress, with none of the pieces definitively fully in place and connected, though elements are getting there, and there is still going to be room for the special skills of Pete Doe, the last presenter, in the field of modeling and fusion, for some time to come.

Set-top boxes

It quickly became clear from Jed Meyer, for example, that on set-top box research, the project has yet to achieve realistic national coverage (though it can access some 2 million STBs at present out of the 60-74% of US households who have them).

Online

By contrast, Jon Gibbs's online brief seems almost straightforward, though here the problem is a microcosm of the overall Nielsen problem: one of integrating a whole range of pre-existing 'products', ranging from the NetView ratings service to Buzzmetrics. Jon predicted that video plus context would outperform 'pure' video and that audience choice would become very important, with the idea of being one's own producer.

Mobile
This cued in Kanishka Agarwal, who pointed out that the mobile is a composite of six different media, with varying existing levels of penetration in the US, with third screen TV  still small, but important for news, weather and sport. With mobile, Nielsen have a programme in place based primarily on a panel of 40,000 lines where they have access to user bills, and this is supplemented by a range of surveys to provide potentially a quite comprehensive mobile measuring kit, though this has yet to be formatted to aid a cross-screen view of video usage.

Three-screen
A key issue for the developing 3-screen world is out-of-home viewing of TV programmes, which the mobile will make more important still. Brian Apter showed how existing integrated measurement, based on a Nielsen-IMMI co-branded service, could supplement home ratings to provide a full rating of the NCAA CBS coverage - in which about a quarter of viewing of the first game of the finals was out-of-home, including visitor viewing; and considerably more of the viewing by 18-34 men. This research will soon expand into a national all-electronic panel, with local services and more detailed demographic and other analyses available, and new panels will be established in other countries.

As Pete doe, the final speaker showed, at present Nielsen can make a good fist of measuring all three screens individually, but is still some way from the comprehensive integrated view that they and their clients might aim for. Meanwhile, there are a number of technical initiatives to fill the gap.