December 15, 2014

Advertising's Arrow Of Progress

One of the interesting aspects of advertising that we have explored from time to time is whether we should think of it more as art or science.

With the growth in the use of mathematics, metrics, and data, it certainly appears like certain aspects of advertising are becoming more "scientific."

However, I am not convinced that advertising as a whole is any more scientific than ever.

From a practical standpoint, there is one factor that clearly differentiates art from science. In science, there is an "arrow of progress." By this I mean, science points in a direction and progresses toward that end.

If you have high blood pressure today, you are more likely to be successfully treated for it than you were 50 years ago.

If you buy a new car, it is more likely to last longer, be safer, work more reliably, and be more efficient than it was 50 years ago.

If you have a personal computer, it can do more things, more effectively, more quickly and more reliably than it did 50...wait a minute. We didn't have personal computers 50 years ago.

The point is, science provides us with technological progress by degrees that builds on itself and improves stuff.

Art, on the other hand, does not have an "arrow of progress." It's not supposed to. Art is about human interpretation -- emotions and aesthetics -- not ongoing improvements. You want to improve on the Mona Lisa? Good luck.

There is no way to talk about whether the work of Roy Lichtenstein represents "progress" from DaVinci. You may prefer one to the other, but to speak about progress is meaningless.

Similarly, is there an arrow of progress from Beethoven to Gershwin? Or Shakespeare to Updike? One may certainly have influenced the other, and styles certainly change, but talking about "improvement" is moot.

That doesn't mean art isn't inventive or innovative. Or that older forms don't influence newer forms. It just means that art moves unsystematically and, unlike science, we don't judge new art based on having "improved upon" old art.

So the question of whether advertising should be considered more science than art rests on answering this question: Is there an arrow of progress? In other words, is advertising more effective than it used to be?

If advertising contains a growing body of useful knowledge that has lead it to become more effective, it should be considered a science. If effectiveness has not improved over time, than it is probably more an art than a science.

Exploring the literature of advertising over the past ten years, one would have to conclude that advertising is less effective, not more. The literature is rife with assertions and research that conclude that advertising effectiveness has diminished over time.

There are certain elements of advertising that seem to utilize scientific principles more regularly -- direct response advertising, media planning -- but there isn't much in the way of conclusive evidence that there is an arrow of progress.

In fact, despite all the hoo-hah over the precision targeting of online advertising, behavioral targeting seems to be only marginally more effective than no targeting at all. And it is not at all clear that this marginal effect is even due to targeting. It may well be that the reason precision targeting appears to be more effective is that the people who are being targeted have been so carefully selected that they are the most natural candidates for buying the product, regardless of advertising.

But even if we stipulate that certain aspects of advertising have become more scientific, I would still contend that the overarching goal of advertising -- the creation of successful brands -- is no nearer to a scientific practice than it was when I entered the advertising business 40 years ago.

Some would contend that the emergence of interactive media, i.e., the web, has led us to a new understanding of brand building that requires electronic co-creating and community building with consumers. The problem with this argument is that a stroll through any supermarket in the country fails to uncover any significant brand of anything that has been built through either online advertising or social media.

From what I can see, despite all the technology we have applied and all the words that have been written, we have uncovered no new generally accepted principles about the nature of brand building or consumer behavior.

Most marketers are still thrashing around in the dark trying to either build a brand or maintain one.

Regardless of the growing veneer of scientific processes, there is no arrow of progress that has helped us understand how to create more successful advertising.



Bob, what do you think about modes of measurement? What are the best ways to measure things? Where does the measurement end?
For example, pharma drugs, many of these tests that prove their effectiveness I think have to be called into question, especially when the tests are funded by the companies selling them.
What I'm basically saying is, are cars safer now? and if they are... has this led to more drivers...and as a side-effect of safer cars meaning more cars on the road, has it made it MORE dangerous for pedestrians...thus one could argue the bottom line is that cars aren't safer. and there has not been any progress. everything 'feels' like we've basically gone/going-deeper into the rabbit hole of modern life and fucked ourselves in our butts so hard, that we can't feel it any more.

dave trott said...


That's great: gave me a whole new angle on Bernbach's quote:
"Advertising is fundamentally persuasion and persuasion happens to be not a science, but an art."
William Bernbach

Jim said...

I do think Byron Sharp's work has added a lot to the understanding of how brands grow.

James Buchanan said...

The improvements you quote are all pretty absolute can be measured in their own terms, whereas the the improvement required from advertising is relative. As culture get's noisier it gets harder to get peoples' attention.

The fact that ads get any attention at all (and some still manage to become national conversations) in a world with so many other distractions suggests to me that some improvement might have happened. That would be hard to show in data though.

As an aside, I don't think the science / art distinction helps in advertising - clearly needs both, often together. That wasn't your question though.

James Buchanan said...

Sorry, "...are all pretty absolute AND can...."

Sell! Sell! said...

I concur.

Tom said...

Part of the problem is we don't have any proper scientists in Advertising.

A proper scientist would realize that there is million things that make someone buy something, and last touch attribution ignores 999,999 of them. It's literally like giving all the credit to the till person at Tesco for selling BOGOF on Holsten Pills to a drunk.

A proper scientist would also realize we get causation wrong. If I like " coca cola" on Facebook, a sensible person would realize that I should happen to buy a can of coke, it wasn't the act of liking Facebook or an ad I saw on Facebook that may have made me buy it. Call me crazy, but people who enjoy coca cola are both more likely to like the page on facebook and buy it, but it ain't down to Facebook.

One day soon when we stop dicking around talking about "big data", the name for what in the 1990's we called "data", we will note that not only do we know loads, and loads more, but we'll realize that most of it shines lights into how much we don't know.

The bit of illumination we do have, will show how fucking terrible most ads on digital platforms behave, the only response they get being at best being ignored or at worse, making people angry as they try to find the (x)

Jason Fox said...

Indeed. As some malcontent on Twitter once opined, there really is a formula for great advertising. Of course, it contains nothing but variables.

Conor Blake said...

Another thing that makes "progress" in this sense nigh impossible is that people get conditioned by what they're exposed to. An ad that might have been very effective 50 years ago could quite possibly be white noise today.

You can't have a formula because every previous input can influence the present. "X" yesterday might not = "X" today, so it's always a moving target.

MangaTherapy said...

I think some of the brands I've seen that have grown online tend to come from fandom (comic books, sci-fi, etc.) interests, but they had offline events (with support from big brands that build their empire traditionally) to build that brand. So yeah, I guess the argument is a bit moot.

Alec Painter said...

Working in the digital media business currently, I see the biggest obstacle to achieving any contribution to brand building is that most advertising treats digital as an interruptive, short-term sales channel.

I'm of course stealing some words from Steve Miles of Dove Unilever, who was able to contribute to some extraordinary brand-building by starting in digital and then integrating all of those efforts in one cohesive push everywhere else.

Generally speaking, the folks working in digital are hooked on the crack pipe of the immediate sale and the pseudo-science of digital attribution and have lost sight of the arrow of progress. But some do get it.

Jeffrey Summers said...

Very, very few!

Steve Jay said...

Isn't a brand really just advertising that is successful in spite of the inability to completely measure it? Measurability makes brand and irrelevant concept.

Tim Orr said...

My frustration stems from the fact that even those things we have learned are often ignored by practitioners of the "art." Colin Wheildon taught us much about typography effectiveness, Jakob Nielsen about web techniques, and the whole world of direct marketing about so much. But try to find a marketer nowadays who has any respect for direct or who even has heard of Wheildon!

It's not so much we don't know how, as that we refuse to learn about the tools, materials and methods we have. The great artists of the 19th century were required to reproduce the "academy" work before they could go out on their own. And that was a good thing. We say until we are blue in the face that we must learn the rules before we can effectively break them, but too many, I think, want to go directly to step two.

Harrovian said...

For me, the explanation for this is quite simple. Advertising probably could and should be, if not a science, then certainly more scientific. But the people involved don't really want it to be. The creators of advertising want it to be art - they like to think of themselves as artists, broadly. And the clients buying the advertising naturally tend to consider it so too - they may not know much about art, but they know what they like, and they judge and buy it accordingly, with a little pseudo-scientific research validation thrown in to make them feel even better about it. The whole enterprise is mostly about subjectivity, and it suits practitioners for it to be that way, not least because a whole load of them would be out of a job if it weren't. More scientific would probably mean less talking about stuff and more getting on with doing the obvious, proven, no-debate things.
It would also almost certainly make for uglier advertising. Take more of the art out and put more science in and you're likely to be left with some stuff that works better but doesn't look so pretty, and on the whole people like to think they're making the world a prettier place. So the instinctive appeal of art triumphs over the mechanistic efficiency of science.

Alec Painter said...

Appending my statement from yesterday while I drink my morning coffee.

I still believe most said above and all of what you said, but I've always taken issue with your supermarket analogy because I feel it suffers from a bit of myopia. None of the brands in a supermarket where born in an era where digital media existed.

There are dozens of brands that have been created solely through the use of digital media (Warby Parker, GoPro, I could go on) and you will probably never find them in a supermarket, much in the way that the supermarket is likely going to go away in a few decades.

The science of digital media is nothing special and the fundamentals of brand building are still the same, but I would shy away from saying that you can't build a brand by using the medium.

Kate Richardson said...

The only supermarket brand I can think of is chobani, although eventually they succumbed to the lure of tv.

LeShann said...

Look at Xiaomi in China. Only digital, only sold online, no retail, they just took over Samsung as the #1 smartphone brand (in the #1 mobile market in the world). But yeah, digital does jack shit for brands... ;)

Charlotte said...

The arrow of progress in advertising points directly to 8 – 12 year olds.
How else would I have known about and purchased Orbitz (drink)? I’d just put the word out on a school yard and watch sales rise.

And for a little education, a link to Oregon State U. engineering page on the other uses for gel beads. Go figure – art = science.