Ten–No, Eleven–Years of Internet Advertising

A few weeks ago, DoubleClick published a report entitled Ten Years of Online Advertising(PDF). It has some interesting stats on the growth of internet advertising over the past decade, and some keen observations about the current internet advertising landscape.&nbsp

In 2004, the report notes:

advertisers in the
U.S. market spent $9.6 billion on Internet
ads, according to the Interactive
Advertising Bureau (IAB) and
PricewaterhouseCoopers (PWC).
That is larger than the whole outdoor
advertising industry, about 80% of the size
of the magazine ad industry and half the
size of the radio ad sector, according to
estimates from their respective industry
associations. Moreover, spending on
Internet ads grew at a rate of 31.5% from
2003 to 2004 (IAB/PWC), compared to
10% for broadcast TV, 7.4% for the
advertising industry in general (Universal
McCann) and 6.6% for the current-dollar
GDP of the U.S. economy as a whole.

Perhaps more importantly, the report notes that 2004 was the first year to pass the bubble peak of 2000 in online ad revenue. And while the 2000 ad splurge was driven by venture-backed dot-coms, the 2004 numbers are driven by traditional advertisers. 30% of all internet advertising in 2004 was done by Fortune 500 companies. But the most significant fact called out by the report is that the growth in interest in advertising is coming at a point when page view growth has slowed dramatically, resulting in a seller’s market for the first time.

I also found very interesting the report’s list of key metrics that advertisers are using to justify their internet ad spending:

  • post-click conversions

  • cost per conversion

  • unique reach of ads delivered

  • average frequency of exposures

  • frequency-to-conversion ratio

  • ad exposure time (rich media)

  • ad interaction rate (rich media)

  • brand impact lift vs. control ad
    (including ad recall, brand awareness,
    message association, brand favorability,
    purchase intent)

  • view-through rate (i.e., delayed visits to
    advertiser�s site without a direct ad
    click-through)

  • share of voice

  • web page eye tracking

  • offline sales lift

  • cross-media-mix econometric modeling

  • I’m not even sure what a few of these mean! Is “share of voice” a buzz-metric kind of thing? And how about web page eye tracking? Not sure how you’d do that except in a controlled experiment, although you could put links at different points on a page, and see which ones get hit. Still, the point is that the web ad industry is getting much more sophisticated.

    The report also calls out the rise of “rich media” (i.e. Flash-based) advertising, not only because of the richer experience possible over broadband, but because it allows for the more sophisticated kinds of tracking listed above. According to the report, 35% of all internet ad spending in 2005 was for rich media advertising. (The cynical part of me has to point out that as rich media is more expensive to create, a boon for ad agencies, 35% of spend doesn’t mean 35% of ads delivered. Nonetheless, this trend is good news for Macromedia. Disclosure: I am on the board of Macromedia.)

    The other major trend is less of a surprise: search engine context advertising represented a full 40% of the ad spend in 2004. And since this is one of the most affordable forms of ad spending, with much less of the money going to develop the creative, and more on delivered impressions, you can imagine just how much of the total advertising pie this is. It’s unfortunate that the report doesn’t give stats for impressions rather than ad spend.

    Yet another key stat that I found fascinating was the assertion that “In its Communications Industry Forecast
    Report in 2004, merchant bank Veronis
    Suhler Stevenson concluded that for the
    first time in history the larger share of
    media revenue came not from advertising
    sponsorship but directly from consumer
    spending, such as satellite and cable TV
    subscriptions, home DVD and videos and
    Internet access.” As customers are using new technologies, from pop-up blockers to the ad-skipping potential of digital TV, media companies are making substantial headway with the subscription revenue model.

    Finally, in a nod to the future, the report gives a shout out to blogging, with a claim that 8% of all internet users now have a blog. And they remark on Chris Anderson’s now famous “long tail.” Much as Dan Gillmor sums up the rise of participatory journalism with the phrase “the former audience”, in his book We The Media, we may be seeing a blurring of the boundaries between advertiser, media property, and consumer.

    As part of its discussion of the growing power of the consumer, the report included the following heartening news:

    Edmond Thomas, chief of technology
    the Federal Communications
    Commission, bluntly warned attendees
    to an AAAA breakfast this March:
    �Your challenge is to stop being
    annoying. You�re almost forcing
    regulators to get involved.�
    P&G�s Stengel was a bit less harsh
    telling the same body more or less
    same thing last year: �All marketing
    should be permission marketing.
    marketing should be so appealing
    consumers want us in their lives.
    We should strive to be invited into
    consumers� lives and homes.�

    Unfortunately from my pov at least :-), the report persists in one error, giving HotWired credit as the first site to offer advertising on the WWW. HotWired did indeed pioneer the now-ubiquitous rotating banner ad in October 1994, but O’Reilly introduced the web’s first ad-supported site, the Global Network Navigator, more than a year earlier, in September 1993 (after a beta launch in March of 1993.) GNN was sold to AOL in June 1995, and didn’t survive the sale by more than a year, alas, but it prefigured everything from web portals (it preceded Yahoo!) to web advertising. It was the first commercial site on the WWW. Here’s the now mostly forgotten back story, starting with the breathless Wall Street Journal article that ran in August 1993.

    I still remember when I came up with the concept of web advertising. We had published The Whole Internet User’s Guide and Catalog in October of 1992. This is the book that introduced many people to the WWW — there were only 200 web sites worldwide at the time we featured it in the book. The book, and in fact, the early development of GNN, preceded the Mosaic web browser, the software that really put the web on the map. In fact, one member of the NCSA Mosaic team claimed that the team learned of the web from a piece of O’Reilly junk mail touting the book. The book included a catalog of internet services, which we turned into a proto-web portal using Pei Wei’s Viola toolkit and pioneering graphical web browser. (This is the same Viola that has featured recently as prior art in the Eolas patent lawsuit against Microsoft.) Dale Dougherty came up with the idea of supplementing the catalog with an online magazine, featuring news stories about new sites, web technology, and the people behind the new medium. What we lacked was a business model.

    I was trying to think about how to support the service, and at that moment, my copy of Infoworld arrived. I suddenly realized that there was a whole class of information products that were free, but supported by advertising, and that GNN could emulate that model. But there was a catch: the National Science Foundation’s “Acceptable Use Policy” (AUP) prohibited commercial activity on the internet. So I asked Steve Wolff, who was the NSF’s guy in charge of internet oversight about our idea at a conference in January of 1993, and he said “our charter is to support research and education, and if there’s anyone who fits that bill, it’s you guys, so go for it.”

    So we did. There were so few web sites that we hosted most of the initial advertisers as part of “resource centers” on the site. The actual ad placement was for small graphical buttons at various places in the service, that led either to the resource center or (as there were more external sites), to companies’ web presences. My idea was that the internet would enable not “ad clutter” — we can indeed thank HotWired for that innovation — but rather that it would allow content-rich, in-depth product information. I was imagining a yellow pages in which you could click through to product brochures, an ad model that would circumvent the “bingo cards” requesting additional product information that accompanied many vertical market magazines.

    And as it turns out, that vision did in fact come true. We’ve been blinded by the rise of intrusive popups to the fact that the real revolution in internet advertising is the radical increase in the availability of product information. What is a site like oreilly.com, or any of a million other corporate sites but an advertisement? Those countless banner ads — they are the barkers outside the door, trying to bring customers into the carnival tent where the real action is happening.

    After the WSJ article hit, we had hundreds of angry letters and emails. But when I pointed out that the advertising on the WWW was different from outbound advertising via email (now known as spam :-), because you were hosting the data on your own site, and visitors came to you, sentiment turned around immediately, and the internet advertising era began.

    By the way, the very first person to cut a check for internet advertising was Dan Appelman of Silicon Valley law firm Heller, Ehrman, White and McAuliffe. He paid $5000 for the privilege of setting up the first company sponsored section on GNN.