Lies, Damned Lies and Ad Predictions

Forecasters Send Us All Into a Tizzy, Yet They're Wrong as Often as Right

By Published on .

NEW YORK ( -- We're not sure whether Nostradamus ever made a veiled prophecy about DVR penetration, but if so, he'd probably have been more spot-on than media forecasters as recently as 2003. That's the year Forrester Research estimated that digital-video recording devices would be available in 25.7% of all U.S. households by 2007.
Brand Identity
Illustration by Joe Zeff Design, Inc.
Many predictions by analysts are more aspirational than rational.

More Info:

The dog ate my homework
Predictions gone awry

Yet when Nielsen released its commercial-ratings data earlier this year -- on the eve of the first upfront designed to accommodate the device responsible for the slow death of the 30-second spot -- the research company pegged overall DVR carriage at 17%. That's the number Forrester predicted we'd be at for 2006.

(Forrester has since scaled back its assessment of DVR carriage for 2011 to be in half of all American households, a number it predicted in 2003 that we'd hit by 2009.)

As any media researcher will tell you, the concept of using benchmark reports and forecasts is intended to serve as less of a crystal ball and more of a hype-driver for emerging technologies and market trends. Even Forrester analyst Charles Golvin admits his prognostications in areas such as DVR penetration were more aggressive than the market ultimately demonstrated.

But he also blames service providers and telecoms for not adapting technologies at a competitive rate -- a result of the Telecommunications Act of 1996. "The telephone companies were reluctant to aggressively build out their infrastructure because of fears they would be enabling competitors to capture their share of a market," he said. "They had initially assumed [forecasts] would justify their expenditure, but it wasn't until the regulatory climate changed that it was actually removed from their consideration."

Still, it's easier to launch an iPhone or a Wii at a speedier rate than the emerging trends of yore. "The first e-mail was sent in 1969, and look how long that took to catch on," said Brad Adgate, senior VP-research at Horizon Media. "The first TV broadcast was in the late '20s but that didn't become a commodity until 30 years later. So we're actually moving a little faster."

Although there are more resources and proven examples to gauge the adoption of a new technology or product, the variables surrounding them are what present the biggest challenges for media analysts, said Mr. Adgate. "The point is more that something's going to grow, but how much is really hard to say," he said. "Generally, in the first five to 10 years there's a thought that ... the price has to come down and there has to be some sort of consumer acceptance. And it takes a lot for that to happen."

Even media consumption has become a gray area for media-ologists. Veronis Suhler Stevenson recently reported media usage per person is down for the first time in a decade, despite initial forecasts to the contrary. Leo Kivijarv, VP-research for PQ Media, which conducts Veronis' annual industry forecasts, blames the terrorists.

Not off all the time
"In the 2004 edition, we were a little more optimistic that time spent would continue to grow during the forecast period due to the impact of 9/11, when consumers were spending more time on media," he said. "We predicted a 2.1% increase in 2006 over 2005, because time spent with media went up dramatically in 2002 and 2003."
Forrester analyst Charles Golvin
Forrester analyst Charles Golvin

But just like the Weather Channel occasionally gets the percentage of precipitation right on an otherwise cloudless day, not all forecasts are completely inaccurate. Veronis was almost on the money in 2003 when it predicted spending on broadcast TV would increase 4.5% by 2007 to $49.5 billion. By 2006, $49.14 billion had been spent in broadcast-TV outlays, Veronis says, a 4.6% increase since 2001.

But the growth of sectors such as in-game advertising and mobile has been predictably unpredictable. In 2005, Veronis said mobile-marketing spending would reach $604 million by 2009, despite only taking in $105 million the year prior in 2004. Research firm Ovum, London, thinks the business could balloon to nearly $1.3 billion in 2009, despite pegging its total outlay for 2006 at $150 million. How's that for a dichotomy of projections?

Video-game advertising, too, was tricky for researchers to forecast in 2001. Jupiter Media predicted video-game advertising and in-game spending would reach $527 million by 2005, yet Veronis recorded only $178 million by the time the year rolled around. Video-game growth, like mobile marketing, has been difficult to track due to shifting subscription models.

In the dark
Don't expect any magic answers when it comes to future ad models, either. Mr. Adgate said he wrote an article in 1985 in which he predicted the 30-second TV spot would be 15 seconds by 1992. The forecast was based on the seven-year rate at which 60-second spots had been cut in half because of tightening of inventories from the Olympics and presidential-election coverage.

"We're still not there," Mr. Adgate said. "Now anything goes. I'm sure there are studies you could do on 42 seconds vs. 1:17. ... If people give you five different areas to predict, there's always two where you're never quite sure."
In this article:
Most Popular