CPMs rising?

So, this has been bothering me for a couple of months. I figure it may well be a figment of the data, or I am abusing the data. But I don't know, so I'll put it up... comments welcome.

This is a graph of Implied Average Online Display Ad CPM, 2006 through Q2 2009 (left axis, thick blue line.) Implied Average CPM is ad spend divided by impressions.


The right axis and thin black line are impressions in millions, as per Thursday's post. This seems to show that as display ad impressions fell in 2008, ad spend did not fall as fast. For this to happen, CPM must have increased. This is both not what has happened, anecdotally, and is hard to believe in this era of expanded access to non-premium inventory. But I hate to think I believe the data when it confirms my preconceptions and then disbelieve it when it doesn't.

Are average display CPMs really nearing $7?

Anyone know what's going on here?

Sources: Ad spend--TNS. Impressions--Nielsen Online. Both purport to be display only.

Ad impressions by month

More data. Nielsen Online's ad impressions per month, via Clickz.com (paid CPM display ads only*.)

First, the raw data.

Broken out by category, sorted by change from August 08 to today: largest decline on top (software), largest increase on bottom (telecommunications.)

Change in four largest categories, and total, since February 06 (Feb 06=100.)

And percentage of impressions by sector. This one's a bit psychedelic.



* Per Nielsen: "Nielsen Online, AdRelevance service uses a proprietary methodology for estimating online advertising expenditures and only takes into account image-based technologies and advertising sold per CPM. Above data does not reflect house advertising activity, strategic partnerships between publishers and advertisers, or text units, paid search, sponsorships, email, units contained within applications (e.g., messengers and pre-rolls) or performance based advertising. "

Better than Panglossian

Andrew Goodman did not like my post Eliminate Advertising. He says "in an attention economy, can you logically even conceive of no advertising? Not even remotely." He doesn't say why not.


But, to be fair, I also don't believe there will be a time when there is no advertising. In fact, I fear quite the opposite. I fear there will be plenty of advertising, and none of it useful.

An investor in my last company--who, whatever else you might say about him, was quite smart--told me once that the company would be successful in direct proportion to the value it created for the consumer. We thought about this a lot as we built the business, and that was no easy thing, because we were in the lead gen space.

And this is the problem: our incentives were not to create more value for the consumer. The consumer wasn't our customer, the advertiser was. This is the problem for all advertisers.

I think advertising is bad. I also think it's good. I guess, in the end, I think it's the worst possible system, aside from all the rest. Pre-internet, how else could companies let customers know what products were available to them in a reasonably efficient way? Producing a message and distributing it was a scale effort; it was only effective en masse. So the cost of delivering the message had to be borne by the advertiser.

This, though, meant that the consumer had to determine if a given ad could be taken at face value. The elaborate game played between legitimate advertisers, trying to signal quality, and the others, who aped the legitimate advertisers' messages in order to ride on their coat-tails, led ads further and further from actual information-delivering devices*.

This Summer I read The Economics of Attention by Richard Lanham. Lanham argues that in an attention economy what matters is rhetoric: style wins out over substance. He argues it quite convincingly. He spooked me, saying that there is no way to have an open attention economy without rhetoric gaining the upper hand.

In this world the best way to decide which product to buy is to listen to two competing producers argue with each other. Like opposing lawyers in court presenting their case to you, the jury. Or like two politicians putting up TV commercials during election season. Neither of these is an efficient way to come to a rational decision, of course, but what way would be better? Lanham views this result as not only inevitable, but the natural order of things. The scientific mindset, that there is some underlying truth, is to him an anti-Hayekian attempt at top-down control, perpetrated throughout our Western culture, from Plato through Feynman. Against this he says: there is no truth, there is only opinion; let each have their say and you decide who you think is more persuasive.

Call me scientific, but I think that some products are objectively better fits with certain people than others. But I am also a Hayekian, and I don't believe that it's right to tell people what products are right for them. I do believe that we can build tools to help people find products better. Goodman and I can agree on this.

Too many people take advertising as it is for granted, though, rather than seeing today's environment as a bit of a historical anomaly. Look at the chart of ad spend per capita above** (this is in real dollars, btw.) The hockey-stick here, starting in the 1950's, was driven by the emergence of mass media; the increase in access to consumers***.

Now we are witnessing, at the same time, both the apotheosis and the destruction of mass media. The apotheosis--reaching every consumer whenever the advertiser wants--has been realized more fully than ever before. And the destruction: the consumer no longer needs to rely on getting their product information from three TV channels, a monthly and two weekly magazines, and a daily newspaper--they can get all the information they want at their own instigation. Both courtesy of the internet and its concomitant radical reduction in the cost of communicating at personal scale.

Apotheosis: the sophistic element of advertising will expand enormously; I think it already has started to. Destruction: the consumer will get the facts on products in places other than ads; I think they already have started to.

As a result, the slow shift of advertising away from think towards feel will become complete. Within a generation, I predict that any ad that tells the truth will be conveying an emotion and that any ad citing a fact will be a con.

If this is so, then what good will it do us as an industry to target better, to buy more efficiently, to have better-converting landing pages? Yes, we need those things and can sell those things short and even medium-term. But, as an investor and some-day-again-entrepreneur, I also want to think longer-term and bigger. Longer-term, companies and consumers will still find each other through a chaotic sea of information, but it won't be what we think of today as advertising. It will be through allowing consumers other ways of determining the truth, and using other, more grass-roots, means to convey context.

I hoped someone smarter than me would read the blog post and say "I know how to do that" and then go do it. That's why I wrote it. I still hope that. There is work to be done and better ways of doing what advertising does poorly now. Goodman calls me a utopian. I'll accept that.

----------------
* This, if you think about it, explains a lot about the advertising agency business, including why good creative is so difficult to make.
** Sources:
(1) Real and nominal GDP, GDP deflator, and population: Louis D. Johnston and Samuel H. Williamson, "What Was the U.S. GDP Then?" MeasuringWorth, 2008.
(2) Ad spend, Douglas Galbi based on Coen numbers.
*** This is not to say that consumers got no benefit, just that the increased benefit was not the driver. In other words, I believe that the rise in spending per capita was accompanied by a decrease in advertising efficiency above and beyond diseconomies to scale.

Not yet Enough

I was reading Skidelsky's critique of Keynes' "Economic Possibilities for Our Grandchildren". In Economic Possibilities, written in 1930, Keynes asked

What can we reasonably expect the level of our economic life to be a hundred years hence? What are the economic possibilities for our grandchildren?

He concluded that by the year 2030,

Assuming no important wars and no important increase in population, the economic problem may be solved, or be at least within sight of solution.

Because economic growth makes us wealthier, at some point people would have enough--"the absolute needs...satisfied in the sense that we prefer to devote our further energies to non­-economic purposes"--and would cut back on hours worked.

This seems reasonable (if a bit far from the problems we face today) but Skidelsky notes that, despite significant progress in the developed countries towards what Keynes viewed as enough, we are not working less.

...We are two-thirds of the way towards Keynes’s target. We might therefore have expected hours of work to have fallen by about two-thirds. In fact they have fallen by only one-third – and have stopped falling since the 1980’s.

This makes it highly improbable that we will reach the three-hour working day by 2030. It is also unlikely that growth will stop – unless nature itself calls a halt. People will continue to trade leisure for higher incomes.

Skidelsky says the reason is that

The accumulation of wealth, which should be a means to the “good life,” becomes an end in itself because it destroys many of the things that make life worth living.

Now, I think Skidelsky is a genius, both in his biography of Keynes* and his more recent book on Keynes' renewed influence, Keynes: The Return of the Master. But I think his acceptance of Keynes' definition of enough--about eight times the average income of 1930--is the flaw, not human nature.

It must have seemed to Keynes that incomes eight times higher than the then-current incomes would be an enormous amount. Today the average US household income is about $68,000**. Imagine if the average household income were north of $500k. Everyone would certainly then have enough, wouldn't they?

This is probably how Keynes felt. And it's true that we've become wealthy in terms of what people thought they needed in 1930. In 1934--soon after Keynes wrote Economic Possibilities--food, clothing and shelter consumed 76% of household income, on average. In 2002-2003, these expenses were only 50% of household income***. (And, god knows, we are consuming more food, housing and clothing than we were in 1934, so our well being has increased more than these numbers indicate.) Far more households can now afford the basics they need to survive, and far more households have much more money left over after they buy these basics.

What have we done with all this new wealth, this extra income? Why haven't we, as Keynes expected, cut back our working hours, and thus our incomes? Why haven't we realized that we now have enough? Skidelsky thinks it's because of our paucity of imagination.

But maybe instead it's because once we had food, shelter and clothing, we realized that we needed more. We've moved up the hierarchy of needs, from physiological to safety. Now that we can, on average, afford the physiological needs, we are spending on health and education. In this light, the enormous increase in the costs of these services might be the desirable result of our ability to finally afford them.

Or, at least, start to afford them. Our healthcare debate is now dominated by whether we really can pay for everyone to have basic healthcare. Costs have increased at jaw-dropping rates, and many feel that we've lived beyond our means for too long already. Even with incomes that Keynes would think were more than enough, we find that we don't have nearly enough for what we now think we need.

But this also points to the solution. Just as we grew into being able to afford the basic needs, we need to grow into the ability to afford these new needs. On healthcare, for instance, the question is not How can we keep the cost down? but, instead, What policies can we enact to be able to afford it sooner? The answer to Skidelsky's question--How Much is Enough?--is that what we have now isn't enough: it would be inhuman to have the means to make peoples' lives better and then not do it, to be able to save lives and then not save them. Instead, we need to think about what we can do to increase our rate of economic growth. What can we do to be able to afford quality healthcare for all, not 100 years from now, but within our lifetimes?

------------------
* Although not so much a genius that I read the full three volumes. I read the 1056 page "abridged" version.
** Median is about $50k and, while the median is more telling about how the average person lives, I think using the average is a better measure when talking about societal income.
*** Source: "100 Years of of Consumer Spending", US Department of Labor Report 991, May 2006.

The FCB Grid and its flipside

A while ago I talked about different views on how advertising works. Personally, I've never believed in a single theory, people evaluate different products differently. They also evaluate products differently depending on where in their product evaluation cycle they are. So, for instance, household cleaners are advertised informatively, while facial soap is sold with brand ads. Car dealers prefer informative advertising while car manufacturers prefer brand advertising.

One of these so-called 'integrative models' is the FCB grid, developed at Foote, Cone & Belding (now Draftfcb) and written about by Richard Vaughn*. This model divides goods and services into four categories, along two axes: the Think/Feel axis, and the High Involvement/Low Involvement axis.


Vaughn makes interesting generalizations about how marketers should address the consumer decision process in each of these four quadrants**. But those are sort of boring, so I overlaid my vague and general idea as to what marketing approaches work, instead.
On the internet, WOM is social media marketing. Direct Marketing is lead-gen and email. And sales promotion is couponing (among other strategies.)

I think this framework is also somewhat helpful in thinking about the various ways to help consumers find the right product (as opposed to selling it to them.)
Personally, I'd like to see a lot more articulation on this last. Helping consumers find the right product will turn out to be a lot more fruitful over the next ten years than figuring out better ways to sell them one.

--------------------------
* Vaugh, Richard (1980), "How Advertising Works: A Planning Model," Journal of Advertising Research, 20 (September/October), 27-30; and (1986), "How Advertising Works: A Planning Model Revisited," Journal of Advertising Research, 26 (January/February), 27-30. Sorry, no link.

** Later refined by Rossiter and Percy: Rossiter, John R., Larry Percy, and Robert J. Donovan (1991), "A Better Advertising Planning Grid," Journal of Advertising Research, 31 (October/November), 11-21. Again, no link. Academic journals suck.

Being allowed to make your own decisions, right or wrong

There was a fairly banal column over at the New York Times last Friday, complaining about being condescended to by a bank "customer service" rep.
“Did you want to add him to the account, or open a separate joint account?” she asked.

“We’ve talked about the options,” I said, giving my husband, James, my secret “not again” look, “but we’d just like to add him to this account.” I smiled pleasantly... “Are you sure?” she pressed on... I assured her that we had considered it and decided to stick to the original plan... She sat back a little in her chair and gave me a half-nurturing, half-scolding tilt of the head. “What would your mother say?”

I went on to read the comments, thinking there would be unanimous annoyance at the bank. Instead, there were an awful lot of people saying that the bank service rep was probably "right" and implying that being right was more important than respecting the customer.

I find this directly comparable to the claims by various boosters that they are doing consumers a service by placing relevant ads in front of them. The idea that someone else is determining what is right for you, based on incomplete knowledge about your situation and your decision process does not seem like an advantage for the consumer. It may sometimes result in a better match between consumers and products, true. But it will always impinge on the consumer's autonomy. I think there needs to be some greater good than slightly better product matching to justify this as a net increase in welfare.

Clearly, if the people advocating the benefits to the consumer of ad targeting believed it, they would instead be advocating better tools to help the consumer choose, not a process that is best adapted to targeting the most persuadable, rather than the best fit.

There are advantages to ad targeting, for the advertiser certainly, and for certain media outlets, but not for the consumer.

Pennywise

I was with my friend Josh today as he pitched a really smart investor on his new company. The investor, as per the script, started picking holes in this and that. Mostly silent to that point, I interjected something to the effect: "this is a pretty cheap option on changing the world."

What is it with VCs?