Serendipity Book

YouTube analytics arrives at last

YouTube: "Today [March 26th] we're releasing YouTube Insight, a free tool that enables anyone with a YouTube account to view detailed statistics about the videos that they upload to the site. For example, uploaders can see how often their videos are viewed in different geographic regions, as well as how popular they are relative to all videos in that market over a given period of time. You can also delve deeper into the lifecycle of your videos, like how long it takes for a video to become popular, and what happens to video views as popularity peaks."

If only you could do the same (or get limited access) for videos that others upload too.

MRS Conference = Groundhog Day?

A colleague protested to me that he was rather disappointed to hear the same old same old at the Market Research Society conference again this year:

"It was a bit like Groundhog Day. It always seems to be themed around self-flagellation over research agencies not getting close enough to client business problems, not getting an airing in the boardroom, not being an attractive option to grads, not having kept up with pace of change in other Marketing disciplines. Wish they'd stop talking about it and just do something about it!"

There were some good speakers lined up but I couldn't bring myself to attend this year largely because of the aforementioned lack of innovation in the programme. Nonetheless I was keen to read this excellent blow by blow account of the conference to see if I had missed anything interesting.

And sure enough it looks as though the conference did have its moments:

1) Andy Dexter tackled the main reason I left the research agency world in his paper where he made the point that "people businesses don’t sit well with volume based business models". The low margin, high volume business model of most major quantitative research agencies is unworkable because it offers researchers no time to think and add value to the data they collect.  The notion of an "insight factory" is clearly an oxymoron in the same way that an "idea factory" is. Andy argued that research agencies and their clients need to "admit that data is a commodity but thinking is not". Not an easy task.

2) Rupert Howell, founder of HHCL and now at ITV, told the tale of the research ITV conducted prior to running 'The Palace'. The research said it would be a hit. It flopped spectacularly.  Rupert suggested that "rather than run away, the research company should work with ITV to find out what happened and how it can be resolved in future". I wonder who he was aiming that barbed comment at?

3) Malcolm White, chairman of the APG and founder of krow, made the astute observation that "planning is currently obsessed with planners and not planning". Such introspection is clearly unhelpful but I remain unconvinced that blogging is to blame as you might expect.

4) Andrew Sharp, once of Initiative and now at PwC, quoted an analysis that claims to have demonstrated that 49% of brands that were created after 1991 were no longer in existence by 2006. The average life-span of a brand created during this period was only 4.1 years and only 11% of brands remained in existence throughout the 15 year period of the analysis.

5) And finally, my comrades at the RLF did their bit on the fringe again this year and even achieved some coverage from the official conference scribes at WARC.

A Millward Brown LINK test for Gorilla?

Meanwhile over on Millward Brown's blog, the ever provocative Charles Frith has challenged the normally unflappable Nigel Hollis to use MB's proprietary pre-testing system to test and improve upon Cadbury's "Gorilla" and make the results public!

So far Nigel has chosen to stonewall a bit by saying that it had already been successfully LINK tested but that he could not confirm or deny the results it got or whether it was changed as a result of the test. Intriguing. Perhaps someone (Mike?) could have a quiet word with Phil Rumbold and ask him to release the learnings to the industry for the greater good.

Charles suggested testing an ad that has not been tested before.  Nigel's comments on an earlier post suggested that he would be up for the challenge. Any suggestions?

Taking on the research orthodoxy

Jason Oke has started an important discussion about the shortcomings of our current use of research in marketing over on his blog.He kicked off the discussion with this deck.

| View | Upload your own

In the comments Jason argues: "The tweaks required to make research more valuable are often minor ones, but ones that are counter-intuitive to people stuck in the old frameworks. They require asking people questions that are less literal, less direct, and sometimes require asking people fewer questions altogether. They also require allowing respondents to have more fun and flow with their answers, giving up some control and direction in the research. These things can seem scary to people who don’t understand the reasons and benefits for doing so. It’s a lot easier, and easier to explain to your boss, to just ask people directly “which proposition statement do you like best.” Unfortunately it’s also completely the wrong thing to do."

Some Free Thinking

First up, a great presentation on the future of marketing from Paul Isakson. He concludes with the statement "modern marketing = making people's lives better":

| View | Upload your own

Which reminds me, I never posted this t'riffic presentation on Insights from Matthew Milan of Critical Mass in Toronto:

| View | Upload your own

And, while I'm at it, I also forgot to post Gareth Kay's splendid presentation from late last year on what makes for a good idea:

| View | Upload your own

As Paul Arden said: "Give away everything you know, and more will come back to you". Start by putting your schtick on Slideshare.

David Byrne: "Analysis is like a lobotomy"

Jon Howard recently posted a fantastic quote by David Byrne of Talking Heads:

"Analysis is like a lobotomy. Who wants to have all their edges shaved off? I’m afraid that everything will get homogenised and be the same. I’m afraid that reason will triumph and the world will become a place where anyone who doesn’t fit will become unnecessary."

Image: my photobooth shot of Barbapapa (not David Byrne)

Apple: "We don't do market research"

Unsurprisingly it appears that Steve Jobs is not an advocate of new product development research. This quote comes from an interview with Apple's 'benevolent dictator' from Fortune earlier this month:

"We do no market research. We don't hire consultants. The only consultants I've ever hired in my 10 years is one firm to analyze Gateway's retail strategy so I would not make some of the same mistakes they made [when launching Apple's retail stores]. But we never hire consultants, per se. We just want to make great products."

However, whilst Jobs clearly likes to give the impression he is flying by the seat of his pants you can be sure that his judgements are founded on some pretty solid knowledge, albeit not the necessarily knowledge that can be bought from a research company. His approach should not be confused with decision-making based solely on intuition, impulse or gut-feel.

Let's not forget how Stephen Colbert addressed George W. Bush after all:

"We're not so different, he and I. We get it. We're not brainiacs on the nerd patrol. We're not members of the factinista. We go straight from the gut, right sir? That's where the truth lies, right down here in the gut."

Image: my photo of the omnipresent Apple ads in San Francisco in 2006

That elusive advertising clutter statistic

Ilya Vedrashko recently went to a lot of effort to track down and validate the elusive source of the oft-quoted statistic that people are exposed to c. 5,000 advertising messages a day. 

In response she received a helpful reply from J. Walker Smith, the president of Yankelovich, who was aware of one of the earliest attempts to estimate this data in the 60s. It makes for interesting reading:

"The oldest such estimate is the one cited by David Shenk in Data Smog. His figure comes from a figure cited in Alvin Toffler’s 1971 book Future Shock. Toffler’s figure came from a conference speech that cited a number calculated by Bill Moran for use in that speech (delivered by his boss) when he was running the research function at Y&R. I know this because I am a friend of Bill’s and he has related this story to me. Bill made a simple calculation. He simply conducted a thought exercise and went through the typical day for a typical person in a typical American big city in the 1960s. How many times would such a person be exposed to some sort of ad, logo or promotion? He came to around 500. It’s that simple, and that’s where this early figure comes from.

He goes on to suggest that all subsequent estimates are likely to have been based on a similar methodology. So my advice would be to take this figure with a pinch of salt kids, it is no more than an informed estimate and I can only imagine that it has become less accurate over time as each subsequent analysis has felt the upward pressure of expectation.

Pre-testing reduces effectiveness

In the current issue of the Marketing Society's journal 'Market Leader', Les Binet & Peter Field make the somewhat contentious claim that, far from improving the chances of getting an effective ad out, quantitative pre-testing actually reduces your chances of success.

Their data shows that ads that have not been quantitatively pre-tested have a 71% chance of being effective whereas those that have been pre-tested have only a 44% chance of success:

Their analysis is based on the IPA dataBANK of 880 case studies. The 'Market Leader' piece is based on their recent book Marketing in the Era of Accountability which analyses the dataBANK in detail.

In the article Binet & Field provide very little to explain this poor performance by the pre-testing industry apart from deriding the overreliance by pre-testers on 'standout' as a metric.

Whilst pre-testing has it's detractors (and I must admit to being less than comfortable with the claims made about most if not all of the pre-testing methodologies that I have encountered), I'm sure that there are factors at play here apart from the use of the wrong metrics. For me the biggest factor influencing the result we see above is the difference in the cultures of the clients that pre-test and those that don't.

Can't count, won't count

The Economist this week reviews a new book called "The Tiger That Isn't" about getting to grips with numbers:

"NUMBERS get a bad press. Almost alone of the academic disciplines, mathematics is one where expressions of ignorance are more of a boast than a shameful admission (imagine admitting at a dinner party that you can't read).

Yet numbers are more important than ever. They are the language of most of science and much of government, two forces that do much to shape people's lives. They are the nervous system of any modern country, marshalled in support of arguments over everything from defence to which diseases should be treated.

Happily for the number-shy, help is at hand. A book about numbers and how to interpret them doesn't sound like interesting bedtime reading. Yet in the hands of Michael Blastland and Andrew Dilnot, respectively producer and presenter of “More or Less”, a BBC radio programme on the subject, that is what it becomes."

Did these numerophobics not watch Johnny Ball? Is everyone afraid of maths?