Robot Last Click

To say that there have been seismic changes over the past 10 years in the ways that we market online is to belabor the obvious. That said, let’s belabor anew as the “00’s” come to a merciful end and we head into 2010. Among the most significant have been the erosion of the walls that existed between the online or digital activities of marketers and those of the traditional or offline variety. In 2000 it wasn’t uncommon at all for a company or brand to have a distinctly separate dotcom marketing team (and even business unit) from the majority of its marketing activities. The notion of silo’d digital marketing has become one that businesses adhere to at their peril as the wisdom of integration and alignment of all marketing channels has grown to become the prevailing wind. In other words, your website had better be telling the same story as your retail locations, you catalogs, your customer service teams, sales teams, as well as PR and brand support. Customers have come to expect your brand’s narrative around value, service or pricing to be consistent across each and every touchpoint they encounter. This is no longer about being a visionary, this about basic blocking and tackling.

Another important change has been the increasing emphasis upon more accurately measuring the ROI of those marketing channels. If businesses have learned anything from their online initiatives, it is that a great many of them can be measured in a way that their offline activities cannot. Of course, the fact that an activity can be measured does not necessarily mean that it should be, at least in the way that we have done so to date. What flows out of this reality is that marketers must apply some manner of metric to everything that they do – online and offline. There is certainly nothing revolutionary about the idea of measuring and applying critical analysis of how resources are spent, but it begs the question of whether businesses are measuring the right things and if those metrics are valid. The beauty of the Google paid search machine is that you get a pretty clear picture of what happens when you put money in the slot, i.e., a click costs a dime, it takes 100 clicks to get an action so voila! the cost to acquire is ten bucks. This leads to ROI calculation that is clean, relatively accurate, and aligns neatly with the silo’d activities of ten years ago. Unless you are a pure-play (online or offline) and market in only one channel, this model can lead to inaccurate metrics at best, and poor decision making around resource allocation.

If only it was so simple to measure the investment in marketing collateral, tradeshow displays, online and offline display advertising, direct marketing efforts, customer retention and lifetime values, search marketing, broadcast, social media, and so on. Truly integrated and multi-disciplinary marketing requires more thoughtful measurement models. Relying upon “last click” attribution today makes no more sense than using a print publication’s circulation and pass-along rates did when the magazine and newspaper business was still healthy. The challenge remains to blend the various aspects of a comprehensive marketing mix and emerge with a more holistic view of what is spent and what that resulted in.

Steve Kerho of Organic reported some interesting findings in a recent post on developing an “analytics ecosystem” model for multi-modal campaigns across the web. Paid search, display (banner) ads and branded website activities showed some inter-relationships that confirm what many marketers have “known” but have had difficulty demonstrating. To wit, paid search performs better when a user sees display ads prior to clicking through a search ad; branded sites converted at much higher rates when display ads were viewed in prior sessions, but after too many display impressions performance fell off. Clearly, utilizing a last-click attribution model around your paid search campaigns risks undervaluing concurrent campaigns in different vehicles. The temptation to discard every activity but paid search requires the development of proxy metrics and blended or weighted analytics to more fully capture ROI. The long story short is that being able to model and manage the data will be a minimal requirement for marketing in the coming years. Marketers that can develop or partner with those that can aggregate and segment data will have the best chance of success going forward. Sound familiar?