Research vs Planning

Image credit: http://www.flickr.com/photos/darkumber/

Preamble

I’ve learned a lot from reading and interacting with various blogs. But something has been bothering me for a while now.

What if it is all complete rubbish?

That’s blatantly trolling and patently not true – I read a lot of fascinating and thought-provoking posts. But the extent to which I should implicitly trust their veracity is something that is concerning me.

There are definitely zeitgeist ideas regularly repeated and regurgitated in the blogs I read. They sound sensible but what are they based upon? Thought? Fact? Data? Opinion?

As ideas are communicated, the content and context changes. This form of “peer review” is in many ways beneficial, but facts can be distorted through misreadings or assumptions.

As a quantitative researcher I deal with hard figures and base my findings upon these. Assumptions are should be anathema to what I do. However research methodologies are far from flawless and one of the more problematic attributes is that it represents behaviour or attitudes at a particular point in time.

This is only of partial benefit to strategists and planners, who need to be able to predict trends. This therefore requires a level of speculation and anticipation, but they need to be based on something. But what?

I read the blogs of many successful planners and strategists, and so they must be doing something right. But to what extent should the prevailing themes of the blogosphere be treated as gospel?

In fact, this is more of uncertainty over “social media experts” than strategists per se, but why spoil a good title…

The problem with speculation

Image credit: http://www.flickr.com/photos/lolita8fotos/

This post/rant was prompted by Pew Research publishing their Future of the Internet study.

“Responding to an invitation to participate in an online survey, 578 leading internet activists, builders and commentators submitted their ideas about the impacts networked technologies may have on world societies by 2020, with an additional 618 stakeholders also participating, for a total of about 1,196 participants sharing their views”

In fairness, I haven’t read the study – it was the principle behind it that got me thinking. What were these predictions based on? As experts expressly invited, you’d imagine that their speculation would be based on some form of study and research, but to what extent should all evidence be footnoted to let us objectively assess their predictions?

Now while they are ultimately speculating, speculation from some people is better than that from others. As Mark Cahill said

“I see Social Media experts who are continually spouting the mantras of the lexicon, yet when you look at their “about me” pages (if they even have their own sites) you find they’ve never built anything.”

Despite some bloggers being well known outside of the internet, a blogger’s identity can be carefully constructed to emphasise certain areas and mask others.

On the Internet nobody knows you're a dog

The distortion effect

As ideas spread, there is the danger of turning peer review into an echo chamber. Mark Cahill also says:

“Are you really expanding the body of knowledge, or merely parroting the Internet A-listers who catch your fancy?”

What was a snappy soundbite can quickly become a buzzword and then a cliché. How many times has the word “conversation” been used in vain since The Cluetrain Manifesto was published ten years ago? How many times did we hear the word “tribes” before Seth Godin’s book came out, and how many times have we heard it since?

Social media “rockstars” aren’t all characterised by a limited vocabulary, but it is a noticeable trend.

Just because I repeat a quotation – and possibly append it to a pretty picture – does it make it correct? Is there meaning or truth behind it? What is the original context, and how does it stand up to intellectual scrutiny?

A quote can be found to back up any argument, no matter how tenuous. How do we know that the selected quote doesn’t contradict the prevailing wisdom? Chomsky does it all the time, so why can’t we?

While this exchange on the We Are Social blog takes quotes from fiction rather than from a thesis or paper, it is nevertheless a great illustration of what I am attempting to relate:

Media CzarTo quote Inigo Montoya, “You keep using that word. I do not think it means what you think it means.”

Faris Yakob“When I use a word,” Humpty Dumpty said in rather a scornful tone, “it means just what I choose it to mean – neither more nor less.”

A house of cards

Image credit: http://www.flickr.com/photos/gibbons/

Scientific knowledge extends through building theories on top of other theories. In astrophysics, very few of these can actually be proved. So if a new discovery is made that contradicts the prevailing wisdom, then the body of knowledge has to effectively regress to the point of proof.

Is our speculation in danger of being built on shaky foundations? While there is little danger of a paradigm shift that mirrors the current financial crisis, there are often missteps. Speculation frantically repeated becomes hype.

Is Twitter the new Facebook or the new Second Life? Who knows? What I do know is the amazing resilience of newspapers to continue to shape the cultural agenda. Internet society is still very much on the fringes. The various trend sites on the web point to interesting but ultimately isolated examples of new behaviour. How many times have they actually been prescient in predicting new behaviour crossing over to the mainstream? Nassim Nicholas Taleb is now regarded as a genius because he anticipated the financial crisis – should all the financial strategists and risk assessors that missed the problem retire in shame?

Snappy quotes and pretty pictures should support the evidence. They shouldn’t be the evidence.
Themes must be built on accumulation of evidence and observation. A quote or a picture may be a better way to convey the thought, but that doesn’t discount the use of appendices or footnotes to support an argument.

Speculative insight

Strategists produce (or, at least, should produce) insights. Or as Rory Sutherland more accurately puts it, ideas.

“What is the difference between an idea and an insight? My own first stab at a definition would be this. That an insight is a sudden and potentially valuable revelation concerning what is; an idea is a sudden and potentially valuable revelation concerning what might be.”

Ideas should be supported by insights, and insights should be supported by evidence. But are they?

Richard Huntingdon believes that insights come from four areas – within, “real people not respondents”, academics and “weird shit”. Academics and a surfeit of real people can produce documented evidence borne out by real behaviour, but the other two? No matter how well informed or intelligent the person might be (and my impression of Richard Huntingdon is that he is both), they still suggest speculation.

And as Nassim Nicholas Taleb has pointed out, we are pretty useless at predicting the future. Among the reasons listed in this fascinating blog post include

  • “Hindsight bias: Past events look more logical, causal and correlated than they really are… It was almost entirely good luck.
  • Vulnerability to “noise”: We are such pattern-seekers that we look for sense in the firehose of information, and find it when it isn’t there
  • The induction problem: The past always seems deterministic. We expect the future to be a continuation of the recent past… So we constantly overreact to the “latest” news (quarterly earnings reports).
  • The survivorship bias: We tend to see only the winners/survivors and lose track of the large numbers of losers.
  • The anchoring bias: Our perception of where we are is biased by where we have come from. If we have a million dollars after living our lives poor, we think we are rich. But if we once had two million, we feel poor, a failure.

Because of these biases, we are, he argues, very poor at assessing risks. I would say we are equally poor at assessing opportunities because of these same biases.”

Where’s the research?

Is research the answer? A mixture of quantitative analysis and qualitative ideas that can be extrapolated to predict the future?

If we believe in the Donnie Darko theory of time travel and path dependent lives then sure.

But I don’t.

Going back to what Rory Sutherland says

“Most conventional research, by coating everything in a patina of rationality, completely fails to uncover the baser (and hence deeper) motivations behind human behaviour.”

Crowdsourcing opinions of future behaviour won’t produce accurate results. We are good at understanding our current behaviour, but not so good at understanding our current motivations for that behaviour. So how on earth can we know what our future motivations will be?

The synthesis

Image credit: http://www.flickr.com/photos/panamaxel/

We therefore need to combine both strategic thoughts and research. For all the evidence we may accumulate, it still needs that eureka moment to anticipate how it relates to the future. These ideas are still not guaranteed success, but it may get us closer to the truth than pure guesswork or pure facts.

As Jeremy Bullmore wrote in his absolutely brilliant “Why is a Good Insight Like a Refrigerator?”

“Conventional market research, professionally conducted, can paint an invaluable picture of the immediate past; but companies also need help in forging their futures… To generate hypotheses you need to speculate: you need to progress from the known to the unknown. But you cannot paint the future in the colours of the past… The origins of an insight are usually to be found in numbers. That’s how we know an insight to be more than airy whim; that’s how we know it has substance; that it can be tested and replicated.”

We may not be able to make predictions with 100% accuracy, but we can evaluate our past predictions. An idea or insight needs to be both provable and measurable. To stand up to scrutiny, it needs to be able to be pulled apart and assessed. The alchemy of inspiration and perspiration.

Conclusion

Planning should be based in research. But equally research has plenty to learn from the strategy side. Data that is only accurate at a single point in time dates quickly – researchers can extend the life of their work by searching for these ideas that move our industries forward.

That’s my opinion anyway. But what’s my opinion based on?

Conjecture. In this post I have made several arguments with carefully selected illustrations and quotes. Where is the data and the proof that my opinion is valid?

Is this just how blogs work? Are blogs just arenas to speculate and should I not read too much into the content? After all, would any client sign off a campaign idea based solely on what someone read in a blog?

Or perhaps I have just missed my calling? This blog is packed full of febrile speculation – what sort of researcher am I?

sk

The cost of giving it away

I am one of a declining number that likes to read a Sunday newspaper.

Recession notwithstanding, I am also one of those people that tends to struggle more in terms of time than money.

Therefore, I generally only have time to read one newspaper a week. The choice of newspaper is effectively zero-sum. I choose one newspaper; the others miss out.

I’ve deviated from that choice in recent weeks. Whereas I used to pick the Observer without fail, a lazy Sunday prompted me to give the Sunday Times a go.

And I enjoyed it. So much that I bought both newspapers again the following week. With time constraints restored, substantial amounts were left unread.

I therefore need to make a choice between the two titles.

And my choice is likely to be dictated by the quality of their websites. Both the Observer and Times offer the majority of their content online in an ad-supported free access model.

But rather than an excellent website causing me to buy the print edition, an excellent website may cause me to forego the print edition.

While print and online may complement, they also duplicate and cannibalise content.

If I am paying for a premium model, I want the greatest improvement in utility to justify that.

This example points to a problem with the Freenium model that I have.

It doesn’t work in perfect competition.

It works for companies like Flickr because Flickr stores my photos and logs my activity. Utility and the cost of switching increase the more I participate.

Newspapers don’t reward relationships (aside from getting the answer to the previous days crossword). So in each transaction, the additional utility in the premium model needs to be justified both against the free version and the competition.

Where (premium, competitive) newspapers are of equal quality, hikes in utility are dictated by the quality of the (free) website.

An inferior website equals a greater hike.

And so the loser in the pitch for my pocket may be that which has invested the most in their website.

Does this mean newspapers need to sabotage their websites in order to increase the value of their premium products? Such as bringing back walled gardens or keeping the best content offline? Henry Blodget thinks so.

Me? Newspapers aren’t my forte so I will resist the urge to speculate. But it raises an interesting question about their ongoing viability in a converged world.

sk

Image credit: http://www.flickr.com/photos/flavio_ferrari/

Reblog this post [with Zemanta]

Links – 22nd February 2009

Some of the things I’ve read over the past week and would recommend:

  • A thought-provoking article in the Atlantic on the future of TV. It argues that TV’s USP is immediacy. While there are still cultural reference points via TV, scripted shows will increasingly see TV as just another distribution pattern. TV will therefore move to concentrate on news, current affairs, live reality shows and sport. This makes sense to me given my research – TV excels at events which are essentially DTR-proof, and the most popular shows online are dramas and comedies that can be viewed at leisure and shared/discussed asynchronously. However, I would argue that successful scripted shows still need TV as that anchor point for mainstream cultural crossover.
  • Ana Andjelic has a great post on our general failure to accurately predict the future. Not only does she argue that a lot of campaigns will fail, but also that our limited perspective means we will often follow the same patterns (potentially of failure)

sk

Reblog this post [with Zemanta]

Green button advertising approaching launch

Channel 4 have announced that they are going to be marketing upcoming shows through the green button service on Sky. Sky’s own forays into green button – originally planned to launch last summer – are imminent.

The green button service is an interactive feature whereby viewers can bookmark content as they view (in real time or recorded). An advert that is interacted with – along with associated content (extended cuts, behind the scenes etc) – is downloaded onto the DTR hard-drive to be watched at a later date. It is effectively advertising on demand. This differs from red button services, which immediately transport you from the content you are watching to the interactive area.

Will it work? The idea is initially counter-intuitive. People choosing to watch more advertising? But we know from the viral/spreadable media successes that viewers can choose to watch adverts.

That is online, where people can remix and repurpose content. Will viewers interact to the same degree on TV? An article in the New York Times says that the market isn’t yet ready for internet TV due to cost, reliability and questionable demand.

But interactive services aren’t the same as having full internet capability, and viewers do seem willing to experiment. Figures from Sky show that more than 93% of digital satellite households pressed red to interact with their TV in 2007 – 16% interacting with adverts.

Although the most viewed interactive advert in 2007 was Cadbury’s Gorilla the campaigns using red button tend to be response-led, with relative success measured by requests for vouchers or for further information. Sky said that 40% of its interactive campaigns are from car manufacturers, who are able to measure sales conversion from the ads.

The appeal of green button appears less about direct response and more about branding. It is essentially viewing advertising as content consumed for entertainment. This is not going to be suitable for all brands or categories, but offers an interesting challenge to companies seeking to broaden their involvement in content marketing and storytelling.

Not only will advertisers have to convince viewers that their content is worth watching and interacting with, but with an on-demand service they also have to move from impulsive to considered consumption. I may see a potentially interesting ad and bookmark it, but will I choose to go back and watch it later?

Downloading content also contrasts with the current trend of streaming. Whether it is putting everything in the cloud, or Spotify emerging as a potential challenger to iTunes’ dominance, owning is partially being supplanted by streaming/renting. Unless I can actively edit or mash-up an advert, is there any benefit to having it stored on my hard drive?

Green button appears to be in direct competition with the Youtubes and microsites that facilitate streaming. Youtube already offers advertising-on-demand, with people able to interact with, share and comment upon advertising. The environment may not be suitable for all brands, but it is cheap. Microsites and branded areas aren’t’ so cheap, but the entire experience can be micromanaged. Green button services are going to have to find a USP that differentiates them and justifies a price premium.

That function may be targeting. I’ve already covered targeted advertising in detail but demographic information could be the pull. If Sky are able to integrate their Skyview panel information with green button, advertisers will know exactly who is viewing and interacting with their content. And that information is valuable.

It remains to be seen whether green button services can make an impact in the period before television and the internet fully integrate, but I anticipate some innovative case studies emerging in the field over the next couple of years.

And not just those looking to prove their environmentally friendly credentials by using “green” advertising.

sk

Image credit: http://www.flickr.com/photos/barbietron/

Is TV advertising responsible for Apple’s success?

I was in a meeting a few days ago where Apple was described as a company that had retreated from large-scale TV advertising, despite TV advertising being responsible for its success.

I disagreed at the time, and remain pretty sure that this is a fallacy. Am I right?

Buttressed by Wikipedia and recent “25 year of Mac” posts, my broad perception of the history of Apple runs like this

  • Set up in the 1970s to moderate success
  • Macintosh launched with ads in cinema and during the 1984 Superbowl (watched by 97m) – initially sells well and ad is regularly cited as one of the best of all time
  • Computer market slumps and Steve Jobs is fired in May 1985
  • Incremental success for the rest of the 1980s
  • Windows 3.1 and – more importantly – Windows 95 take the PC to the next level and nearly kill Apple
  • Jobs comes back in, ends most of the product developments and places his faith in the iMac
  • iMac becomes a success and a design classic
  • iPod launched – the aesthetic of white earbuds and “1,000 songs in your pocket” become ubiquitous
  • iTunes overhauls an outdated music distribution system
  • iPhone brings touchscreen technology and simple web surfing to the masses
  • Halo effect of the Apple range boosts the computers – Apple is currently the number 4 computer manufacturer in the US
  • Jobs’ ill health and the rise of netbooks raise questions over Apple’s continuing success in the computer market

To my mind, TV advertising doesn’t play a particular big role in this rise, fall and rise of Apple. There have been iconic campaigns – 1984, Think Different, the dancing silhouette – which have contributed to the success. But they have not driven it.

That is the Cult of Mac.

The iPod may be mainstream, and the iPhone may be getting there. But Apple is not traditionally a mass market company. Their computers appeal to a niche audience. They may be the no.4 manufacturer, but the choice is PC or Sony. It is not Dell, Acer or Mac.

However this niche audience is passionate. They follow. They promote. They evangelise. They attend(ed) Macworld every year and hang on Steve Jobs’ every word.

That community is what has driven Apple’s success. Apple concentrate on the product – usability, design, experience. That leaves the marketing to the community. Alan Wolk has an interesting post on this – good advertising can accelerate success, but a decent product to win over the public is vital. In the case of the iPod, the evangelism changed an industry.

TV advertising has made plenty of products successful – from Hofmeister to Barclaycard to Cillit Bang. But Apple isn’t one of them.

Unless someone like to correct me?

sk

Image credit: http://www.flickr.com/photos/sigalakos/

Reblog this post [with Zemanta]

Crowdsourced consumption and product development

We are all consumers. We choose what to consume. But we also choose how we consume it.

This may be completely different to how the inventor anticipated, or expected, usage. It is the law of unintended consequences.

Contrary actions may prevent a product or service succeeding. But equally, an innovative use that builds on the original concept can be game changing. As Cory Doctorow points out in the article, unintended consequences include such inventions as

Even in a planned economy, inventors can’t dictate how their product or service is used. The community decides. No matter how tightly controlled a campaign is, the product owner cannot dictate outputs.

Similarly, companies cannot dictate perceptions or experiences. Products facilitate actions, but the consumer decides what these actions are. Faris Yakob argues that “A brand is a collective perception in the minds of consumers” while Justin Porter says that “Designers do not create experiences, they create artifacts to experience”

The internet and social networks may enable individual customisation and organic social constructs but these aren’t new phenomena. The community has always had power.

To jump on the bandwagon, I’ll use Twitter as an example. Media coverage of Twitter concentrates on either celebrity uptake, or its position within breaking news stories. These are popular expressions of Twitter use, but were they anticipated?

One of Twitter‘s co-founders says that it was initially conceived as a text dispatch service. Did they foresee that Twitter could be used for, among other things

  • Microblogging
  • Socialising publicly
  • Socialising privately
  • Linking
  • Echoing
  • Syndicating
  • Following
  • Polling
  • Tracking
  • Searching
  • Reporting
  • Announcing
  • Promoting

I suspect that some of these actions would have been considered. But not all of them.

Friend or Follow is a simple tool that can highlight the different ways that people use Twitter. Promoters will have more followers, followers will be fans of more people, and socialisers will have a high proportion of friends. Each person will use Twitter in a slightly different way.

The Twitter founders accepted and embraced this. And took it a step further. Recognising that business development is constrained by their own imaginations (among more prosaic factors such as finance), they opened up their API to outside developers and effectively told them to “go wild”. Individual consumption is supplemented by devolved development.

These tools may have contributed to the growth in popularity as those not convinced by the core offering may be sold by a new application. Some of these – such as Summize – have even been incorporated into the core offering.

This movement takes the concept of community power to the next stage. Product development has joined consumption, perception and experience in being co-opted by the crowd.

But the consequences aren’t wholly positive. Twitter had to buy Summize. With a closed API, they could have introduced a search function organically with minimal cost. A multitude of functions could create a paradox of choice, or confuse the core offering (while perceptions are individual, a company needs a core offering as an anchor). Not to mention the Firefox levels of bloat that could ultimately occur.

By effectively trusting the masses to create Twitter apps, the company also leaves itself open to nefarious or incompatible activity. This could be fake services phishing passwords, or a service that – for example – could promote fascist behaviour.

This makes the path of product evolution far more unpredictable than within a tightly controlled environment such as Apple (though their iPhone apps store does partially relinquish this). How can Twitter remain confident in their product experience. Is some level of control necessary to ensure a coherent thrust, or can the community be relied upon to promote the positive innovations and marginalise the negative?

I don’t know the answer to this, but it also poses some questions regarding usability testing. Is the only way to facilitate this in real-time through public beta testing, rather than closed-shop research initiatives? A balance between organic, crowd-sourced improvement and publicly known missteps need to be found.

To conclude this wildly incoherent post, I remain fascinated by the ability to appropriate not only imagery but functionality. Whether Twitter, or the burgeoning RFID-enabled internet of things, it is an area producing a myriad of innovative activity.

sk

Image credit: http://www.flickr.com/photos/heyyu/

Reblog this post [with Zemanta]

Links – 15th February 2009

Things I have read in the last week that I would recommend:

Henry Jenkins has begun serialising his white paper on spreadable media – If it Doesn’t Spread It’s Dead. Part 1 – on media viruses and memes – and Part 2 – on sticky and spreadable – are both fascinating

Andrew Scott argues that Google Latitude is a Trojan Horse into social networking with the ultimate aim of combining location with context/mood

Robin Grant provoked an interesting discussion around conversational marketing with his post Learning to Speak Human

On the Digital Design Blog, companies are told that actions speak louder than advertising, and therefore Brands should do

Ana Andjelic riffs on Kevin Kelly’s post on sharing and copying by pointing out the differences between economies of scale and economies of scope

Adriana Lukas re-iterates the distinction between advertising (information) and Advertising (disruption)

Sean Howard argues that the IAP2’s spectrum of public participation is backward. He believes that empowerment and trust need to come first; as an input, not a result

Clay Shirky on why micropayments won’t save publishers

Lovely Charts is a web application with an accurate description (though some might quibble that for basic users, the use of the singular would be more reflective)

sk

Reblog this post [with Zemanta]