To register, get added benefits and unlimited access to articles, Become a Member. Already a Member? Sign in.

Yes, Web Advertising Scales -- If You Measure It Right

Why a Lack of Standards, Even for Basic Demographics, Hurts Us Online

By Published on . 6

Scott Knoll
Scott Knoll
Surprisingly, some people continue to trot out the tired argument that digital advertising doesn't scale, because the web offers too many choices.

There's no question that online advertising has failed to grow in proportion with the amount of time people spend online. But attributing that fact to "too many choices" is like blaming Bulgaria for the Second World War. Yes, Bulgaria was aligned with Germany -- but it was hardly the driving force.

Eyeballs always range ahead of ad dollars. A lot of people were listening to the radio before advertisers caught up. The same thing happened with TV. And the same thing is happening with the web.

Still, just waiting "for the advertisers to catch up" oversimplifies things. There's a subtler dynamic at work. Not all advertisers are staying away. Approximately 60% of direct-response advertising has migrated away from snail mail and TV to the web. That's an historic shift.

Clearly, not all advertisers are shying away. Only the ones with the biggest dollars: the brands.

The extent of the direct-response shift, in fact, only underscores how far behind the brand dollars really are. Only about 4% of the pure brand-advertising dollars -- just $4 billion -- have moved online. The other 96% is staying where it's been: TV, radio, print, highway billboards.

Brand advertising is way behind DR advertising -- and it's not catching up.

A closer look shows us that it makes sense for DR to be so far ahead. Right now, most of the online-ad innovation focuses on direct response. Data targeting, ad exchanges, demand-side platforms, buy-side platforms, lookalike modeling, retargeting -- it's all about DR.

And it's not just technological breakthroughs brands are waiting for. It's also a breakthrough in etiquette.

As hard as TV is to measure, if you buy space on "The Bachelorette" in an effort to reach women 18-34, and you don't reach them, ABC will give you a make-good. They'll make you whole. On the web, that doesn't exist. If you don't reach those women online, all too often you don't even know that you didn't reach them. So you can forget the make-good.

How can we expect people to invest big dollars in a medium that doesn't tell them whether their dollars are being spent well?

Those of us in the industry love to tell ourselves that the web is the most measurable medium ever. And it is -- for DR measures. When it comes to measuring reach and exposure, though, the web is worse than TV or radio.

That's partly because there are no standards, even for basic demographics. Panel-based measurement tools like Nielsen and ComScore are simply not robust enough for the web. On TV, when you have large audiences with a few hundred choices, a sample size of 12,000 households can be enough to tell you what's going on. But the panel methodology, which the web has borrowed from radio and TV, breaks down when applied to millions of fragmented audiences exploring billions of choices.

The current panels these companies use, for example, can measure user demographics with statistical relevance only on sites that have at least 20 million uniques. Hardly any sites get that much traffic. If you applied that standard to print publications, even The Wall Street Journal would be too small -- it wouldn't benefit from statistically relevant audience data.

Another way our lack of granularity hurts us online: On espn.com, the panel methodology forces advertisers to treat the audience that hangs out on the NBA section as though it were the same as the audience that obsesses over the PGA.

A lot of smaller venues try to get around this by surveying their readers, and reporting the results to prospective advertisers. Talk about a solution that doesn't scale! Imagine if print publications didn't have the ABC audit bureau to confirm their audience. Imagine if advertisers had to rely on each publication's description of its audience, each using its own nonstandardized survey -- and had to take their word for it. You better believe that would stifle advertising growth.

There are no standards for measuring reach and exposure. Until there are -- until a company like Procter & Gamble can run campaigns on 30 sites, and can compare each campaign on a granular level to be certain they're reaching their target -- it'll be hard to justify a big online spend. Until a company like P&G can feel confident, and can confirm after the fact that its confidence was well-placed, it will keep most of its dollars offline.

And it will be right to do so, because there's no measurement out there.

Brands aren't stupid. They know their audience is spending more and more time online. They know 30-year-old mothers no longer give "General Hospital" their full attention, or even part of it. They know those moms are on the web, playing games and interacting with friends and strangers. They just don't know how to reach those moms there, or whether they might have reached them, because we haven't shown them. At least not yet.

ABOUT THE AUTHOR
Scott Knoll is general manager of Datran Aperture.
In this article:

Comments (6)

Read These Next