ComScore announced Monday some new proof on the efficacy of view-through as a key performance indicator for Web advertising.

It can’t feel good to be the head of marketing at SK Telecom lately. With even the Wall Street Journal proclaiming that, in effect, you’re squandering shareholder value, the board meetings must be pretty uncomfortable. Safe to assume that the recruiters aren’t calling either.

At the AAAA Media Conference in 2005, P&G’s Chief Marketing Officer Jim Stengel prognosticated, and I’m paraphrasing, “In 10 years we’ll be evaluating media buys based on engagement.” Within days, “engagement” was the hottest buzz word in media. Councils were formed, papers writ, models developed, products built, all specifically to provide insight into media’s new most important metric.

Last Friday, Jim Stern wrote in this blog about the difficulty of getting people to undertake the measurement process — and the difficulty in getting tags (pixels) up. We’ve all run into these issues. I recognize that the current time is such that getting your work done every day is the biggest issue. But it is also a chance for change. And if we can vote to change a whole country, we can certainly consider rebooting our tagging systems and creating something newer and simpler.

Kim Johnston, vice president of global sales and marketing operations at Symantec Corporation, gave a killer presentation at the eMetrics Marketing Optimization Summit last week in Washington D.C. The ah-ha moment for me and many in the audience (the response was audible) was a diagram she showed of all the marketing campaigns running for just one of their products. Kim said she looked at all the promotions that were not delivering measurable results and cut them. She then cut all the projects that were not being measured. The new diagram showed almost half of the programs had vanished. The result? More than 20% lift in response for the same dollar.

In many of my contributions to this column, I’ve written about the efficacy of panel-centric measurement. Two weeks ago, I talked about the promise of panel-centric hybrids, writing specifically about comScore’s Video Metrix service, which incorporates publisher beacon data into our Media Metrix panel measurement. Today I want to take a step back and ponder a question that has been on my mind lately, as conversations about panel-centric, site-centric, and hybrid models intensifies: what, exactly, IS a panel?

There is, once again, an increase in the market of CPA deals. This generally happens in a down market, when less inventory can be sold on a CPM basis due to market weakness. Mark Kvamme of Sequoia opined at a AAAAs management lunch in San Francisco last week that CPA would be THE sales metric of the future. We don’t agree that CPA will be the only sales metric of the future. There are two reasons for this: First, we can sometimes make better deals with CPM-based buys and second, sites cannot see the full picture from a metrics standpoint and CPA represents only a single dimension.

The ups and downs of the market this past week have brought many companies to a screeching halt when it comes to spending (and productivity). It has been quite a rocky road recently, but there is a silver lining to this dark cloud when it comes to the role of the Web analyst.

The Google Ad Planner was announced with much fanfare at the Advertising Research Foundation’s Audience Measurement Symposium 3.0 conference in June. Consumers of audience measurement data generally expect a reasonably high degree of transparency and disclosure with respect to methods, but the Google methodology remains something of a black box, whether we’re talking about the source of data collection, the nature of the data (person or machine, server side or client side) or the demographics.

The formal announcement of WPP’s acquisition of TNS has me wondering, who is this good for (other than WPP and some TNS shareholders)? What are the metrics for success in this deal? Clients, promised the benefits of synergy (a.k.a. “effectiveness”) and efficiency, may find that objectivity and differentiability are more important metrics. Let me explain….