Was GM's Facebook ad failure GM's or Facebook's fault?

Government Motors’ cancellation of their $10 million Facebook advertising campaign just days before Facebook’s more than $100 billion IPO on May 18 has triggered an industry debate:

Does GM’s killing of the campaign because their paid ads “had little impact on consumers” mean that there are systemic weaknesses in Facebook’s paid advertising component (which could raise serious doubts about the IPO)?

Or does it mean that GM is just continuing its long corporate tradition of shortsighted, inane decision-making (a tradition which antedates Facebook by decades)?

You can make a good case either way, and, in fact, some smart people have.

The case for Facebook’s fault

Within hours of our report, Larry Kim, founder of WordStream, e-mailed to say that as a medium for paid advertising, Facebook is a poor second to Google Display Network. He cites several reasons:

  • Facebook, his research has determined, has a substantially lower average click through rate (CTR) than Google — 0.051% vs. 0.4%. While true, this statistic alone fails to take account of two things. First, it’s an average, and the way averages work is that if your head’s in the freezer and your feet are in the oven, on the average you’re at a comfortable temperature; it’s possible to substantially beat the averages on the basis of factors other than media selection. Second, Google’s eight-to-one CTR advantage is at least partially offset by Facebook’s exposure — 1 trillion page views per month vs. Google’s 180 billion monthly ad exposures. Multiplying those figures by the respective CTRs, you average 510 billion click throughs per month with Facebook vs. 720 billion with Google. As a ratio, that’s a substantially smaller difference. (And, for local or regional Richmond advertisers, probably an irrelevant one, because the target audience is too small to be measured in billions or even millions.)
  • Facebook ads have short — by implication, shorter — shelf life. The CTR declines by about half within two days of the ad going live. No comparable figures are supplied for Google.
  • Facebook ads can’t be seen on smartphones and tablets.
  • While Facebook offers one-size-fits-all size and layout formatting, Google offers a wide range of different sizes and positions. That variety can help smart advertisers make good messages stronger, but it can’t save lame ones.

The case for GM’s fault

While GM’s slamming on the brakes with Facebook ads, Ford’s going pedal to the metal.

“We are doing more advertising on Facebook,” says Ford’s director of marketing communications, Matt VanDyke, “and it is a growing and critical part of our media mix.”

It should be, because GM’s historic nemesis has been using paid Facebook advertising with great success.

When they switched their Explorer advertising from Super Bowl television to Facebook, they saw a 104% jump in shopping. Their Fiesta Movement Facebook campaign produced tons of qualified leads for test-drives at dealerships, and the test drives produced an amazingly high conversion rate.

And recently, Forbes reports, Ford combined paid advertising and content on Facebook in a “customizer” app for the Mustang, which included sponsored placement of a video on the logout page. “We got one million views in one day,” VanDyke said.

So Facebook ads can work when you do things right. According to an unnamed Ford source, GM doesn’t. As Ford tweeted:

It’s all about the execution. Our Facebook ads are effective when strategically combined with engaging content & innovation.

A perfect example of where Government Motors went wrong is their “Plant A Tree” campaign for Chevrolet last Fall. The idea was, you download an app to plant a virtual tree, which Chevrolet will match with a real tree in a national park (which sounds about as necessary as importing ice cubes to the Polar Cap).

Unlike the successful Ford campaigns, this has nothing to do with buying or driving cars. One can only imagine that the reason they put together this exercise in green political correctness was either an outbreak of mass insanity or massive arm-twisting by their biggest stockholder (the United States Government, Barack Obama, President).

According to the unnamed source, the problems went beyond the campaign’s content (or lack thereof): The campaign was set up as a one-time interaction “instead of giving users a reason to come back over and over again.” What’s more, once the app was built, there was inadequate investment in channels to get it actually seen. “You can’t take an ‘if you build an app, they will come’ approach,” the source explained.

Shooting the messenger?

It may be that GM’s Facebook advertising “had little impact on consumers” because they have a brand problem instead of just a Facebook problem.

Among Baby Boomer, Generation X and Gen Y women, a crucial segment of both Facebook’s and GM’s audiences, not one GM product is on TrueCar.com’s list of the ten most popular car brands with women.

GM ranked 12th out of 13 brands in this year’s Consumer Reports automotive scorecard.

And as recently as a little over a month ago, they recalled some 50,000 vehicles. Planting virtual trees won’t overcome that.

Splitting the baby

The answer to the whole GM vs. Facebook debate lies somewhere in the middle, but probably closer to Facebook’s side.

Facebook may not be the most effective or cost-efficient advertising medium. But smarter media buying, in and of itself, never overcomes a bad message or bad products. It just exhibits their flaws to more people, sooner.