• by August 30, 2001

On Monday, there was an interesting article in the New York Times about measuring online adverting. It cites measurement - once thought to be the "biggest strength of Internet advertising" - as its "most conspicuous weakness". It got me to thinking, "what if we never had any of the metrics that we have at our disposal today?"

How would this have affected the industry? Would it have been larger or smaller than it is today? Would it have been held to the same standards and measures that are used to gauge the effectiveness of television, radio, print and out of home advertising? Would we have been more or less proactive in terms of innovating and implementing commonly-used research tools in our interactive campaigns?

Many questions...all leading to perhaps the same discussion point: Are the metrics we have at our fingertips today a blessing or a curse? Have they harmed or benefited the growth of the industry?

They're both. A blessing in terms of giving us the ability to learn so much about our current and prospective customers; A curse in that it has also been the proverbial rope around our necks.

The irony is that online metrics gave us what had been so sorely lacking in the other media - accountability. We acted like greedy children gaining access to the pantry cupboard - we ate too much too quickly and then got sick.

Take the clickthrough. In the first years of online advertising's existence, we've either been stuck in the CTR quagmire or we've raced to embrace "new" and "advanced" metrics such as "viewthroughs". But have we even perfected the click? Do we even understand what a click means?

In a cluttered environment, with more people being exposed to more impressions (at a cheaper rate), a falling CTR is inevitable. But the number of actual clicks on the other hand is more revealing. If the click is the beginning, then the visit is the end and the clickthrough is just the means to an end.

Then there's the frequency issue. Once a person clicks, they most likely will not click again. So naturally, clickthrough is going to be adversely affected, but not necessarily at the expense of the campaign's effectiveness. We recently ran a campaign which yielded below than industry-average CTR's on a particular site, yet that same site was one of the top referring URL's to our client's website (we had to work a bit harder to arrive at this meaningful conclusion, which was that people were making their own way to the site).

In the cases when the intended action goes beyond a visit, we need to be absolutely certain about how we define and measure a "conversion". In addition, we should be managing realistic expectations about what is and what is not achievable based on this definition.

What about branding campaigns? Is there a role for a click in these types of campaigns. Absolutely. Whilst the goal is not to solicit a click, there always needs to be some kind of back-end fulfillment to the online advertising promise. That's just the nature of the medium. The online consumer demands the right to be able to aggressively pursue additional information of value and we in return, need to provide that avenue to fulfill their needs.

We recently measured one of our branding campaigns by concluding that we delivered a given number of high-impact impressions that clearly conveyed our campaign's primary message. In this example, clicks were a bonus, indicating the potential presence of "interest" or even "intent to purchase".

My opinion differs slightly from the Times viewpoint. I don't pin the onus on measurement itself, but rather on those responsible for its collection, interpretation and dissemination. Through back-end analysis, we have the ability to not only understand the way consumers interact with our advertising, but also what happens once they arrive at the intended destination. We need to be employing a diverse mix of advanced reporting and analysis techniques in order to help us achieve this.

So back to the original question: how would we have done without the CTR and similar metrics? Probably quite well using other media as a yardstick. This is where "traditional" research would have come into play. It's also where we would have used a combination of intelligence, intuition and innovation to extract and derive the kind of knowledge that clients expect from us.