Fog Creek Software
Discussion Board




The advertising emperor is naked

This is a bit off-topic for JoelOnSoftware, but it relates to Joel's earlier comment (from a few weeks ago) that advertising doesn't work.  A few people posted messages saying, in effect, that advertising _must_ work -- otherwise, nobody would use it.

Well, today's _New York Times_ ran a good article about irrational behavior in the advertising industry.  This article is consistent with my experience.

I'm certainly not an expert, but the impression I got from working with advertising agencies is that they present a veneer of science.  However, all the focus groups and interviews and questionnaires don't add up to much -- the final decisions are based more on gut feelings and/or office politics than on objective, quantitative models.

Anyway, the _New York Times_ piece argues that advertisers have an unjustified, irrational pre-occupation with the youth market:

<quote>
The Myth of '18 to 34'

[...]

Eighteen to thirty-four: for decades, conventional advertising wisdom has attached the adjective ''coveted'' to this slice of the viewing audience. According to an analysis by the former NBC News president Lawrence K. Grossman, advertisers pay an average of $23.54 to reach 1,000 viewers in that age bracket, versus $9.57 per 1,000 over the age of 35. And since commercial television, whatever else it may be, is fundamentally a system for delivering audiences to advertisers, network executives lose a lot of sleep trying to figure out what will hold fast the slippery attention of people in their late teens, 20's and early 30's. It is, as it has been for 40 years, the principle by which a great deal of our popular culture -- not just TV, but music, movies, radio -- comes into existence.

The odd thing is, there's no real reason for it anymore.

People over the age of 50 account for half of all the discretionary spending in the United States. Proportionally speaking, there are more of them than there ever were, and they are voracious cultural consumers. They watch more television, go to more movies and buy more CD's than young people do. Yet Americans over 50 are the focus of less than 10 percent of the advertising.

What makes advertising an entertaining field of study is that its twin natures -- pop art and dismal science -- are never really reconciled. If the notion of the ''target demographic'' lives on well past the point where it stopped obeying any kind of economic logic, it may be worth wondering how much sound, unsentimental business sense was ever behind this juggernaut to begin with.
</quote>

http://www.nytimes.com/2002/10/13/magazine/13DEMOGRAPHIC.html?pagewanted=all&position=top

Alex Chernavsky
Sunday, October 13, 2002

I think advertising has been of one of the great scams, like UML.  There are the odd Cokes and Nikes who really used the medium well, and the ad agencies used them to tell everyone how much money there was ripe for the taking. 

Same with the Microsofts.  People think there is an enormous amount of money in "technology" and see that the industry lags behind real technology by decades.  Easy pickings?  No, people have to learn how to eat before they can create great things their entire lives.

Greg Neumann
Sunday, October 13, 2002

I had to take an advertising course when I was in college.  The rational behind the 18-34 market is that they are more influenced by advertising.  In other words, 50 years old will buy the same soap, beer, toothpaste, and clothes regardless of what they see on tv.

On the other hand, 18 year olds are very much influenced by advertising.

IMHO, they are on to something.  My father doesn't pay much attention to advertising but I do (and I did ever more at 20).

vanguard
Sunday, October 13, 2002

In the animation world, there is always a huge pressure to target 'tweens' (10-14ish) ((which incidently means animations about high school students, the ones tween look up to)). It isn't because they buy all that much, it is because they make a lot of life time decisions based on advertising. I wonder if the author of that article would also be puzzled at the tobacco industry targeting kids, who account for an even smaller percentage of their market than the 18-34's?

He did skip over the lifetime brand thing a bit I think - sure people are less brand loyal now, but what was that 70%? 70% of Coke drinkers are willing to switch to Pepsi? I doubt it. 70% of Land'o'Lakes butter are willing to switch to Modern Daries? Probably. The real question is what percent of advertised products no longer have a strong potential for lifetime appeal, which I don't know, but I bet more than 30%. Not to say Madison avenue isn't ever out of step with reality, I'm sure it is, but I doubt the implied conclusions of the article are all that accurate either.

I think it is useful to separate the concepts 'does advertising work?' and 'is it being used effectively?'. The article is about the second question, not the first - but I bet this thread doesn't stay that way ; ).

Robin Debreuil
Monday, October 14, 2002

Advertising works at even more basic levels, if I don't know it exists, how can I but it? Without advertising, how do new ideas get to market?
Old products such as coke, butter, chocolate, toilet paper, potato chips, etc personally I couldnt care less, they are all one and the same thing to me, when I'm in the supermarket, price will probably be the only thing that I consider, but then I'm a male, over 40, as mentioned above my skin is very very thick when it comes to advertising, but then as I get older and fatter and my hair keeps falling out maybe I'd be prepared to buy that sports car, with the beautiful woman in the passenger seat, sitting next to the driver, who strangely, is about my age...

Alberto
Monday, October 14, 2002

[I think advertising has been of one of the great scams, like UML.  ]


Just how is UML a scam? Just because Rational touts its process as the end all solution to development does not mean UML should be labled as a scam. Can you describe an application in detail without using UML diagrams? I'd be interested to see how this is done.

Ian Stallings
Monday, October 14, 2002

UML may have its place, but to "describe an application in detail without using UML diagrams" is easy: Just use the source code.  That works for me.

xyzzy
Monday, October 14, 2002

So when you are designing your application, you use source code to describe it to others? Doesn't that require writing it before designing it?

Robin Debreuil
Monday, October 14, 2002

[I think advertising has been of one of the great scams, like UML. ]

OK. let's wander this thread all the hell all over the place... Thanks, I will ;-)

But, to follow-up on this branch of the thread, I agree this comment is wrong. Perhaps it's just incomplete, but as it stands, I do not agree that "UML" itself is a scam. It is just a tool. It can be used well or poorly.

Now, I _could_ buy a statement that "over-hyping UML as being the silver-bullet to cure all your development ills is a scam", were anyone to make such a statement. In fact I just did, because such hype is a scam.

My assessment is that many otherwise good tools and techniques are somewhat "polluted", sometimes to their detriment, by being over-hyped. Hmm. perhaps this comes from the advertising some get, which may mean this comment isn't so far afield of this thread afterall.

Most solutions only work on part of a larger problem, and most solutions depend upon some preconditions being satisfied. Advertising and sales pitches that mis-represent these realities or omit them completely, are the worst kind of hype (and are all too common). And people who listen to the crap and don't apply critical thinking to it are doomed to suffer the consequences. Of course, what often happens is that the solution at the center of the original "hype" gets a bad rap, when the problem is belief in the hype about the solution, not the solution itself.

Without knowing more, I suspect the author of the anti-UML comment has gotten burned somehow by some inappropriate promises made or expectations set about what UML itself can accomplish.

Cheers,

anonQAguy
Monday, October 14, 2002

No, the next sentence I wrote that the Cokes & Nikes were able to use advertising to great effect, implying that UML can be well used too.

I just talked yesterday with a sysadmin who told me that UML will soon make programmers redundant.  (Interesting.)  I think the hype-meter is pretty high on UML.  All I'll say is that it's useful in large projects.

Greg Neumann
Monday, October 14, 2002

Another problem with UML is that in many people's minds, UML (the diagramming spec) and the Rational Unified Process (the development process) are one and the same.

The RUP is unabashed, absolute hype. That taints the useful parts of UML (the diagrams).

Chris Tavares
Monday, October 14, 2002

Greg -

Oh. ok. sorry for misunderstanding what you meant.

anonQAguy
Monday, October 14, 2002

"And people who listen to the crap and don't apply critical thinking to it are doomed to suffer the consequences."

That is exactly the reason advertising gets you though - no one can apply critical thinking when they want to be convinced. That is, when they haven't formed opinions yet, they get a positive emotional or otherwise boost and all the other tricks. Your biases are much stronger than your logic, so critical thinking is mostly bypassed. Once you have decided though, you have a great capacity to resist a message, easiest is a logical one - look at OS's, relgion, politics etc. That is why 10-35 (depending on the product) is such a 'coveted' demographic.

Fooling the world into using a second rate product would probably still fail long term, but usually it gives you time and money to improve until you are better. A lot of tech products start as an average idea plus hype, then grow to be great. Giving your product a good image is the same as wearing a suit and nice hair to a job interview - it is a form of lying but it can get a chance you might have otherwise missed.

Robin Debreuil
Monday, October 14, 2002

"Can you describe an application in detail without using UML diagrams? I'd be interested to see how this is done."

To give a couple of publicly available examples, Don Knuth managed to do it quite successfully for both TeX & Metafont, and John Lions did it for Unix. Do you feel that their descriptions would have benefited from the use of UML?

Andrew Simmons
Monday, October 14, 2002

I used to own a painting business,

I would send out 1000 peices of mail (250 Dollars)

I would get 4-8 responses each netting an average of 1400 Dollars (I would typicall get 2.5 sales)

so 250 invested 3400 dollars returned (minus roughly 800 spent on  materials)

(After the inital 2.5) I would also get 1 or 2 referrals
for another 1500.



how does advertising not work? (granted this was 10 or so years ago)

I think what Joel should think of is advertising will not help with repeat customers, but to get people to try your product, yep it works ...

Daniel Shchyokin
Monday, October 14, 2002

Yeah, Daniel - I agree that advertising works. At very least, it lets people know of the existence of products/services they might not otherwise know of, even if the audience doesn't go right out and buy a particular product based on an Ad's content.

And Robin - you're right. That's how they get you. I agree that the critical thinking would help if it happened more, and that it just simply doesn't.

anonQAguy
Monday, October 14, 2002

Daniel wrote: "how does advertising not work?"

Targetted mailings probably work, as you say.

I'm skeptical of large, expensive advertising campaigns, though -- the ones you see on TV, print, radio, etc.  I doubt that these have a positive return on investment.

See: http://www.techtv.com/news/business/story/0,24195,7798,00.html

Alex Chernavsky
Monday, October 14, 2002

I wonder if advertising guys have discussions about programming?

Engineer and Designer
Monday, October 14, 2002

Ah Alex,

So what people mean is IMAGE advertising doesn't work, and for the most part yes. There is a book by Jim Caples about advertising very good read (his opinion is the same)

Daniel Shchyokin
Monday, October 14, 2002

The problem with Image advertising especially in tech, is that 99.9 percent of the population doesn't know what you are talking about, and only 1% of those that do, can buy your product/service... once again I work in tech with java so I am one of that lucky .1% that does know what BEA is (they advertise on top of taxi cabs here in SF), but I am still never ever going to be in a position to buy one of their products (even though I think its a decent product)!

Now people told me before that  well.. its so consumers recognize the brand ... but that is incorrect too, I will bet it is pretty rare for someone to call there ISP and say something like "hey, do you run your servers on sun? cause if not ..." In most software there the Customers are the VPs of big companies, and there are too few of those for Image Ads!

Daniel Shchyokin
Monday, October 14, 2002

If you don't think advertising works, then you've never heard of Coke, Nike, McDonalds, Ford, Budweiser, etc. etc.

pb
Monday, October 14, 2002

pb wrote, "If you don't think advertising works, then you've never heard of Coke, Nike, McDonalds, Ford, Budweiser, etc. etc. "

Hearing about them isn't the issue.  _Buying_ them is the issue.

Of all the brands listed above, Nike is the only one that I purchase on a regular basis (and it isn't because of their advertisting -- it's because I like the way they fit).

Also, how many ads have you seen for Google?

Word-of-mouth promotion is the way to go, if you're planning a marketing campaign.

Alex Chernavsky
Monday, October 14, 2002

I don't think there's any doubt that on the whole -advertising works. Business spend bajillions of dollars on it. We can't be naiive enough to suggest that huge companies (who only care about money) would make such massive investments based on anything less than a sure prospect...

Advertisers get paid to manipulate our brains - and to some degree, they do. Robin put it beautifully when (s)he said: 'Nobody thinks critically when they want to be convinced.' If only the world could overcome that....

For our industry though - I think people are smarter. People making software decisions are generally shrewder, or they know somebody who they trust to evaluate products for them. I think it's getting harder to market software products effetcively,  because the Skept-O-Meter on our prospective clients in generally becoming much finer tuned...

Gordon Taylor
Monday, October 14, 2002

"I don't think there's any doubt that on the whole -advertising works."

Myself, I have considerable doubt.  People are very good at fooling themselves.  I'm sure that lots of executives are convinced that they're getting a positive return on investment when they spend gazillions on Superbowl ads and such.  But are they really, or are they deluding themselves?

For a similar case of mass-delusion, consider antidepressants.  Remember when we used to think that Prozac was a wonder-drug?  Oops:

http://www.usatoday.com/news/healthscience/health/drugs/2002-07-08-antidepressants.htm

Alex Chernavsky
Monday, October 14, 2002

If Slashdot is always there to catch when Microsoft does something like make fake Apple "switcher" ads, I don't want to know what other companies do.
http://slashdot.org/article.pl?sid=02/10/15/0044255&mode=nested&tid=109

We're in the age of fuckedcompany.com.

Tj
Tuesday, October 15, 2002

[For our industry though - I think people are smarter. People making software decisions are generally shrewder, or they know somebody who they trust to evaluate products for them. I think it's getting harder to market software products effetcively,  because the Skept-O-Meter on our prospective clients in generally becoming much finer tuned]


This is exactly why some companies are targeting the "CXO" level with their advertising and sales tactics. They try to convince the CEO or other higher up that their product is worth investing in, the next thing you know you got to rewrite your latest application to take advantage of the "big ass box of xml" server now residing in your server room. This type of sales/advertising ensures that the people doing the critical thinking are out of the loop. The IT department is surpassed and the CEO lets the flood gates open for a second, squeezing out another million for that PDA integration server.

Ian Stallings
Tuesday, October 15, 2002

"For a similar case of mass-delusion, consider antidepressants.  "

Surely that "case of mass-delusion" was the result of an awesomely effective advertising campaign aimed at GPs, the general Media and depressed people?

Gordon Taylor
Wednesday, October 16, 2002

Gordon, I suppose it depends on what you mean by "advertising".  When a salesman from the Eli Lilly Company visits a doctor's office and makes a sales pitch for Prozac, I don't consider that to be advertising.  When the Eli Lilly marketing machine manages to get Prozac put on the cover of _Newsweek_ magazine (March, 1990), I don't consider that to be advertising, either.

When I say I'm skeptical of advertising's effectiveness, I mean that I'm not at all certain that most television ads, magazine ads, etc. have a positive return on investment.  More-subtle types of promotion probably work well.

Alex Chernavsky
Wednesday, October 16, 2002

Yeah - I agree. I think that advertising is:

"Any concerted effort to imbibe somebody's brain with information that affects their ability to critically evaluate products in favour of the advertiser."

That pretty much covers anything that any marketing department could come up with ;-)  Advertising works, but there's no doubt that some forms are less effective than others...

Gordon Taylor
Wednesday, October 16, 2002

Good advertisers are capable of selling products that nobody really wants, of unproven efficacy, and at inflated prices...

...including advertising!

Max Hadley
Thursday, October 17, 2002

Max -

And you've hit the bottom line regarding advertising right there. Advertising most certainly works - people pay for it, it makes the advertisers money, end of story. Anything else it may do, or people pay think it does is gravy/incidental.

anonQAguy
Thursday, October 17, 2002

*  Recent Topics

*  Fog Creek Home