• by Naomi , Chris Young , March 28, 2005

Since the first online ad was sold in the mid '90s, our industry has made enormous strides and, in my view, is about to pull up a chair at the grown-up table to be served its rightful share of ad dollars. Yet, some things about online never seem to change, such as the debate about whether at the end of the day it is a direct response medium or a branding medium on par (or even better than) TV and print.

Since, unlike most other media, the interaction with end users can be measured by their "response" some have pushed the CPA model insisting that the Internet is only effective if it can provoke an immediate "desired action." Others have shown that because online can provide a nearly limitless amount of consumer information and selling points about brands, that with sufficient reach and frequency, it is a superior branding medium. This argument occurs not only at the macro-Internet level, but the micro-ad-unit level.

I will go out on a limb here and say, as far as online video, it can be both.

As proof, I provide some enlightening statistics from John Kerry's presidential campaign, which at different times asked the Internet to be both a direct response vehicle and later, a branding medium. Early in the race it was imperative to John Kerry and the Democratic National Committee (DNC) to raise money to jumpstart his campaign. The result was a video that did three things: introduced the candidate, asked for e-mail names and ZIP codes (for a variety of targeting objectives later) and solicited a donation. More than 13 percent of the users who clicked on the John Kerry video ads stayed with it long enough to donate money towards the campaign. This is a terrific rate for any kind of direct response.

Was it the creative or the ad unit? You decide. By contrast, similar standard GIF and Flash ads, showed a click-to-donation conversion rate only of about 2 percent.

After the Democratic convention, the campaign moved into its "branding" phase or to voter persuasion. The DNC ran different spots created specifically for the Internet. During this part of the campaign, the percent of video viewed metric became a key indicator of success. Users overwhelmingly viewed a large portion of the ad before clicking through. For example, 59 percent of viewers watched between 18 to 24 seconds of the 24-second spot. Put another way, of the more than 50 million video impressions run by the DNC, approximately 30 million views watched at least 75 percent of the spots.

The DNC ran several ads throughout the campaign, including some units that presented users with multiple videos based on different pressing issues. The first incorporated three TV spots covering three different issues: the first hundred days, health care, and special interests. An analysis of the campaign was able to determine which videos were receiving more play than others providing a guideline to what was resonating with the public. In the final weeks of the campaign, Internet-only video featured key debate footage.

The campaign certainly raised money from online ads earning its direct response credential. But since Kerry did not win the election, does that mean the branding aspect of the campaign failed? That we will leave to the political experts. But the key success metric was that users were engaged by the online video often enough and long enough for its messages to be delivered. Can anyone say with certainty how long TV spots were watched or which ones struck a responsive chord with viewers? The cost to adjust online messaging was infinitesimal compared to changing TV commercials in and out.

Last year's political campaign did not generate the online spend we all hoped and expected it would. But when campaign managers look at how each medium performed for their candidate's varying objectives, I expect online will come to be the predominant medium of choice for future political races.