Friday, August 15, 2014

shoplifters of the world, unite and take over (or: why online advertising is mostly a flop)

Yes, we can do several things at once, but only if they are easy and require little in the way of thinking.

We are mostly reasonably skilled at performing a number of 'automatic' or 'system 1' type processes - chatting to a passenger and listening to the radio while driving, for instance.

However effortful mental activities activities that interfere with each other, multiplying 17 x 24 while making a right turn into dense traffic - the example Kahneman often uses - is much more difficult and should probably not be attempted.

On occasions when I have to interrupt a colleague to answer a question or something, I ask - as they are typing or whatever

'Can you multi-task for a moment?'

'Of course' is the reply.

The subject then either stops what they were doing to listen to my request.

Or continues to type, more or less ignoring me as I speak.

Either way it goes I derive some psychologist humour from the situation.

The truth being that no-one can multi-task particularly.

Magicians and illusionists understand this better than most, particularly those who practice 'close-up' magic of the kind that is often performed at restaurant tables.

One of our favourites is the famous Derren Brown example. In one of his TV shows Derren was able to 'pay' a Hatton Garden jeweller with blank pieces of paper for a 1000GBP watch, by sufficiently distracting the vendor's system 2 with complicated questions about London bus routes.

I was once inspired by this trick and managed to get an Australia Post employee to bag two imported British music mags worth about $50 for free by confusing him with questions about domestic Australian vs overseas stamps.

As I am a pseudo-scientist rather than a habitual shoplifter, I returned to the counter within a few moments and coughed up. The smugness of winning was reward enough.

The term for this is inattentional blindness, and is most famously demonstrated by the famous invisible gorilla experiment.

Inattentional blindness occurs when selective, focused attention towards one task renders us 'blind' to other peripheral happenings around us.

However both mine and Derren's experiments are dwarfed by this example of real world shoplifting skill as reported by ABC news.

A Texas woman has succeeded in stealing $57,000 worth of iPads from various Wal-Mart stores by loading up her trolley with various sundry items, allowing the cashier to scan the ipads, placing these items in her own handbag then informing the cashier that she needs to nip back to aisle 3 for teabags or something then marching straight out of the store while the hapless Wal-Mart employee is busy scanning the rest of the items in the basket.

So we can be blind to the obvious, and we are also blind to our blindness.

This could go some way to explaining the general failure of online advertising versus the continued effectiveness of, less fashionable, outdoor advertising.

It is simply far more easy - extreme creativity allowing - to get a tiny bit of attention from people who are not concentrating on doing something else that is more important.

Tried living in the real world instead of a shell.

But... I was bored before I even began

Friday, August 08, 2014

a note on recognition heuristics and moving from party tricks to business results

Until recently we were unfamiliar with the German psychologist, Gerd Gigerenzer.

It turns out that Gigerenzer and Kahneman were at odds for many years.

This is despite that both researched and studied heuristics and came to about 95% of the same conclusions, however the principle point of conflict appeared to be; Gigerenzer believed in a kind of ‘expert intuition’, whereas Kahneman is more sceptical.

Whereas the Kahneman and Tversky school (ie much of the behavioural economics lobby) principally associate heuristics and biases with human error  (to be fair, in Thinking Fast and Slow DK does say that his view had softened over time, much System one activity can also be described as 'skill'), Gigerenzer assert that heuristics not necessarily cost the decision-maker and are often 'smart'.

These are the semantics that cause the big rifts in Academia. Heh.

What is initially startling about Gigerenzer's ‘Gut Feelings’ (we are reading this just now) is how much of his schtick Gladwell appropriated for ‘Blink’.

In 'Gut Feelings' Gigerenzer references H G Wells, who famously noted "If we want to have an educated citizenship in a modern technological society, we need to teach them three things: reading, writing, and statistical thinking."

To this day we teach reading and writing to children but not yet statistical thinking in any significant way.

Perhaps we won't get much better at avoiding certain cognitive slip-ups until this is addressed.

An example of a typical statistical error is given from the New Scientist magazine.

A Food inspector visited a restaurant in Salt Lake City famous for its quiches made from four fresh eggs. She told the owner that according to FDA research every fourth egg has salmonella bacteria, so the restaurant should only use three eggs in a quiche.


Anyway, Gigerenzer describes a 'smart heuristic' as a tool for making good decisions in an uncertain world, where one sometimes has to ignore information in order to make an optimal choice.

His principal gift to the advertising world is one such smart heuristic - it should be said that this one crops up less frequently than it's usefullness would indicate - the recognition heuristic.

In simple terms, this means that when faced with a choice between something familiar and something unfamiliar, people tend to opt for the former, and this choice is often the best choice.

In other words, it's usually sensible that people should place a higher value on something they recognise over an alternative that is less familiar.

Therefore recognition based heuristics help consumers choose which brands to buy in frequently purchased categories.

Another way to describe this is salience.
Salience in this context, being the propensity of a brand to come to mind in buying situations.

It seems simple but, and this is something you can try at home, when asked to name as many brands in a particular category as possible, people can rarely name more than three or four in that first instant.

After a few moments thought a few more can come to mind, but theres rarely that kind of reflection in a buying situation.

Though when read a list of brands in a given category then people will be 'aware' of a far larger number.

Therefore, unaided salience is an better predictor of the brands people will buy.

And a useful example of a recognition-based heuristic.

As Professor Sharp notes in 'How Brands Grow'; there are many brands that buyers could consider, if they had thought of them.

I've introduced the good Professor at this point for a specific purpose.

I attended the MSIX (Marketing Science Ideas Exchange) conference in Sydney last week.

A fuller review of this I intend to write shortly, but one of the stand out (salient) points was made by one of the speakers - a non-marketing person - Jon Williams of PwC who said that for behavioural economics to have an impact on business/marketing then 'we need to move from party tricks to business results'.

My sense is that is what Professor Sharp was alluding to in an article he published in Marketing Mag this week entitled 'Behavioural Economics is not the big story in marketing'.

Professor Sharp says:

'But what really worries me is that this infatuation with theatrical psychology lab experiments – that show participants in lab experiments can be nudged without knowing it – is bringing back a discredited theory of brand image of besotted/manipulated consumers'

Then goes on to describe recognition heuristics almost exactly.

'Yes, consumers use heuristics to make ‘good enough’, not perfect, decisions. But they use these heuristics because they work rather well. For example, the assumption that higher-priced items are better quality works pretty well because it’s largely true.

Then later adds.

....don’t forget that the main reason your sales aren’t what you’d like them to be is that your brand doesn’t have the mental and physical availability to produce the demand that you would like.'

Many months ago I read a piece by Mark Earls who also expressed some scepticism abut the ability of behavioural economics to paint the whole picture. At the time this didn't really register however there is a picture becoming clearer now that is pointing towards a more unified theory.

What Professor Sharp describes as 'mental availability' - or salience - is the same thing that Gigerenzer would describe as recognition. A major factor in recognition is the sense of popularity - and this is core to Mark Earls's thesis.

These factors combine to create demand.

Fulfilling demand, however, is heavily dependent on 'physical availability' and much of the so-called 'party tricks' of nudg-ery are perhaps best employed in that context, if only from the point of view that even if recognition is a principle driver we will tend to 'satisfice' right at the last second if one of the other couple of brands that are salient present themselves more easily.

Anyway, to round off we were delighted to discover the very first TV ad for the fledgeling VW Golf featuring The Munich Beefeaters Dixieland Band with one Gerd Gigerenzer on the steering wheel and banjo.

As an undergraduate Gigerenzer was a keen trad-jazz banjo player, playing in jazz bands at night in order to fund his university studies.

Wednesday, August 06, 2014

when johnny comes marching home

This Guinness spot from the US caught our eye, from a scientific standpoint.

Perhaps a teeny bit schmaltzy and all-American for Aussie tastes but worth noting it does a few things well.

Well branded right from the off, the product in action using repetition.

The core narrative is, of course, super easy to get, it's 'When Johnny Comes Marching Home'.

Bar 'America The Beautiful' that's as full-on a slice of Apple Pie as one gets.

[Funnily enough the lyrics to When Johnny Comes Marching Home were written by the bandleader Patrick Gilmore, an Irish immigrant during the American Civil War.

Even more peculiarly the song was widely adopted and sung by both sides.]

One responds emotionally (I detected a slight physiological response too, you may also depending on how much of a wuss you are).

I’ve trained myself to let go in these matters and be the punters - don’t fight it, feel it - and duly recorded a nice tight peak-end rule being deftly applied.

Also worth noting that the main creative device employed is in the area of 'personal triumph'.

Johnny makes it all the way home.

If you've perused Karen Nelson-Field's findings on viral (or not) video then you will have noted that the data would suggest that this device is the one that tends to benefit from more shares than others, with the caveat of significant paid promotion to maximise initial reach applied.

Gung ho.

Tuesday, July 22, 2014

the great advertising swindle (actor-observer bias remix)

Advertising's outcomes are notoriously hard to measure.

Which is why in advertising agencies, we love to measure outputs instead.

Agency outputs (ie creativity, ingenuity, technical wizardry and planning cleverness) are far easier to evaluate than the contributions that the work has to actual client business outcomes.

Despite this clients will often clamour for performance based remuneration deals from their agencies.

Some agencies even claim to offer this.

How they can do this is unclear.

I'm oft to remark that if it were possible then Y&R London should still be receiving a performance based royalty from Heinz Beans for the work of their then deputy creative director, the young Maurice Drake, who penned the famous tagline 'Beanz Meanz Heinz' over a couple of pints of a lunchtime back in 1967.

Maybe they are, who knows?

The fact is that the effects of great advertising often take a long time to unfold.

And many other factors other than the advertising will affect brand performance.

(The effects of shit advertising tend to reveal themselves much sooner, of course).

So when, among its peers/competitors, agencies performance can - for the most part - only be evaluated in terms of the creative output, then the agencies themselves are highly incentivised to squeeze the juice out of those outputs regardless of whether those outputs can be said to have contributed to business outcomes.

Even the planners effectiveness awards, known as Effies, are no more precise, for the most part only describe shorter term effectiveness, and are often forced to rely on qual/quant research of highly dubious methodology (such is the ineptitude of our market research cousins, but that's another topic for another time.)

Perhaps you have been following the ad scam kerfuffle over at Aussie media commentator website mUmBRELLA.

If not, then in summary, it has come to light that a few Australian press ads that picked up Lions at Cannes this year benefited from somewhat 'limited' distribution.

While the ads met with the criteria for entry, it has been argued that running one time in a suburban newspaper of little consequence is not fair play.

There's plenty of opinion around this. Several articles have been written, each receiving large numbers of comments.

I'm not going to add further opinion, but perhaps offer some thoughts towards the beginnings of an explanation.

From personal experience I've played in two broad camps.

Both highly incentivised to pursue award-winning activities.

In big network agencies, consistently receiving global and local awards was an absolute imperative.

At [agency X], for example, every Monday morning we would roll in to work and be curious to find out which awards the agency had won over the weekend.

And every significant piece of work seemed to have its creative award entry case study video constructed almost in parallel to development of the campaign itself.

To my knowledge there was never anything cooked up specifically to try and manipulate creative awards, however every piece of work was evaluated internally for its creative award potential, and certain pro-bono work was often considered for the same reasons.

There's nothing hokey about this approach. It's correct.

However, if a major award show came and went where the agency performed poorly in terms of metal then one could feel the pressure to return to winning ways a soon as possible.

In this environment, where success is routine, then winning becomes table stakes.

In smaller agencies or even decent sized indies there is a different sort of pressure.

To level the playing field then these agencies have to look that little bit harder to find the opportunities to generate outputs that can stand up against the outputs of agencies with more resources and better clients.

One could describe the situation as a kind of agency double jeopardy.

Smaller agencies get hit twice. They have fewer, less sophisticated and resource rich clients, who also tend to be less loyal.

This provides the smaller agency with plenty incentive to 'maximise' and then possibly manipulate outputs in order to portray their creative abilities in the best way.

Because no agency is going to attract new sophisticated clients with a portfolio of mediocre work.

By hook or by crook you need to get the goods.

Now, we are aware that the agencies under scrutiny in the Aussie case are DDB, Saatchi's and to a slightly lesser degree JWT. All outposts of big global networks, not small by any stretch.

However, the Aussie market is somewhat peculiar inasmuch as just about every global network has an office in at least two (often 3 or more) of the major cities.

Everyone is scrapping with everyone else.

In the absence of any way to meaningfully measure clients' business outcomes, the industry evaluates itself in the only way it can. Though outputs - and in the arena of award shows.

The volume and quality of new business an agency attracts is explicitly connected to the volume and quality of the awards they accrue.

The more you get, the more you get.

Without making any judgement call on what-is-or-is-not-scam perhaps some clarity comes from knowing this about our own foibles as an industry.

And perhaps it might not be a bad thing to put this years what-is-or-is-not-scam debate to bed and get on with next years award winners.

Because, as an industry perhaps we suffer from a collective actor-observer bias.

When we judge our own agency's behaviour, we are the actors, and perhaps we are more likely to attribute our actions as a response to peculiarities of the situational factors of the industry; than to any general sense of our integrity or lack of.

However, when explaining the behaviour of others (our competitor agencies), we are far more inclined to attribute their scam ads to their overall cheating-bastard disposition rather than to any of the situational factors that influenced us.

Friday, July 18, 2014

stand and deliver

Pop stars talking about advertising. What could possibly go wrong?

Corporate rapper Kanye West for a start.

Consider, his much repeated quote-age from Cannes for example.

“I dream to help raise the palette and raise the taste level of a generation and also be involved with the production and distribution and advertising of that thing everyone’s begging for.”

I have no idea what that is supposed to mean.

Adam Ant, however, knew a thing or two about the role for advertising and its subsequent effectiveness. And preferred to express this in how the work spoke for itself.

'I'm the dandy highwayman who you're too scared to mention

I spend my cash on looking flash and grabbing your attention'

It's not that difficult.

Spend the cash on looking flash.

Then get noticed.

Tuesday, July 15, 2014

shot with your own gun (or the META-return of the texas sharpshooter fallacy)

The sciences of human behaviour show that we all are susceptible to foibles known as cognitive biases.

These are processes of thinking that can sometimes lead us to making less than optimal decisions - particularly in conditions of uncertainty - or more often just relatively benign and harmless routine confabulations.

In the following case, the firm favourite - confirmation bias.

This is our tendency of people to favour information that confirms pre-existing beliefs or hypotheses.

And it's often compounded with a touch of the old texas sharpshooter fallacy.

Refreshingly, from time to time behavioural economists themselves are just as likely as any of us to be caught in these cognitive traps.

In theory, this analysis in Smart Company of how the celebrated Sarah Wilson’s 'I Quit Sugar' books and 'community' have helped a large number of people adopt more healthy behaviours seems to add up.

A slightly edited excerpt follows that links BE staples status quo bias, intrinsic motivation, loss aversion and social proof with positive and negative tension.

Here we go.

'[Wilson uses] positive tension to create anxiety about status quo [and] create an appetite for change.

By pointing to the gap between what’s undesirable now and what is a desirable future you can stimulate sufficient motivation in your customer to be open to changing their behaviour.

Loss Aversion tells us that people are more motivated to avoid loss than seek gain, so you need to work hard to reduce the perceived downside of progressing with you...overcoming negative tension to build people’s willingness to change.

By providing step-by-step meal plans as well as cooking and shopping tips, Sarah reduces people’s fear that a life without sugar is too difficult.

Further, through a thriving blog (engaging over 260,000 every day) and social media presence (over 385,000 followers) Sarah has harnessed a community of advocates and tapped into the behavioural principle of Social Proof.

People new to the materials or program who may feel anxious about whether it will work for them can have their fears allayed by seeing how many others have succeeded. There’s no surer anxiety-buster than knowing that someone else has done it before you.

This analysis is not wrong. We just smiled at it's possible over-theoreticalness.

[And before you start, sure, the irony of that statement is not lost on us...]

Another, perhaps simpler, explanation for the success of the 'I Quit Sugar' phenomenon can be found by applying the Fogg Behaviour Model.

Central to Fogg's theory is that increasing motivation is very hard and almost never works.

Making the desired behaviour easier, even where there is low motivation is generally a much better plan.

For Wilson's wannabe-sugar-free readers there is already likely to be a high motivation to change their behaviour, however because of a perceived lack of ability, it seems hard to do.

In cases like this Fogg recommends the use of a 'faciltator' trigger.

Make the desired behaviour easy by showing people who are motivated how to do it, easily.

Wilson's books, forums, recipes are facilitators, and this is how her empire has been built.

Buy making something that used to be hard, easy.

And branding it well.

Of course, there's more than one way to skin a cat, and the planner joke applies, sometimes what works in practice doesn't always work in theory.

Having said that, we are signed up for the Marketing Science Idea Exchange later this month, at which the author of the piece is conducting a couple of workshops.

In the spirit of idea exchange, hopefully we shall not be sent to the back of the class.

Friday, July 11, 2014

space travel's in my blood, and there ain't nothing I can do about it

We've just finished reading 'Think Like A Freak', the third book by Levitt and Dubner the others being Freakonomics and Superfreakonomics. You may also know their weekly WNYC radio show/podcast.

Their ouvre is at the lighter pop end of the behavioural economics canon, however they can take some of the biggest credit for opening up this thinking to a mass audience.

Either way it's an entertaining read, I got through in about two round trips to Hobart.

There's one little story towards the end concerning the January 1986 Nasa Space Shuttle mission which I suspect that many of you - my planning brothers and sisters - will identify with.

The following is almost as it appears in the book, I've paraphrased/shortened it a bit...

The launch of the Space Shuttle Challenger had already been delayed several times, both nasa themselves and the engineers were getting twitchy.

However on January 28, 1986, it looked like Nasa could finally launch from Kennedy Space Centre in Cape Canaveral, Florida.

This particular mission had drawn lot of interest from the public, mainly because the crew was to include a civilian, a schoolteacher called Christa McAuliffe.

However some unusually cold overnight temperatures in the nights before the launch led the chief engineer Allen McDonald to recommended to his Nasa client that they postpone once more.

McDonald and his team of engineers explained that the cold weather might damage the rubber O-rings that kept hot gases from escaping the shuttle boosters, indeed, the boosters had never even been tested at temperatures as low as forecast.

Nasa pushed back, they wanted a launch.

Years later Mcdonald wrote about this.

"This was the first time that Nasa personnel ever challenged a recommendation that was made that said it was unsafe to fly.

For some strange reason we found ourselves being challenged to prove quantitatively that it would definitely fail, and we couldn't do that."

So the advice of the engineers was ignored and the launch was officially back on.

McDonald - and his team of experts - had been overruled.

In fact when Nasa came and asked McDonald to sign off on the decision to launch he refused.

His boss signed it off instead.

So the next morning, Space Shuttle Challenger took off as scheduled.

73 seconds later Challenger blew up in mid-air, killing everyone on board.

A subsequent enquiry found that explosion was caused by a failure of O-rings due to the cold weather.