Search Marketing as a Metric

We marketers like to consider the entire SEM (define) process as a form of advertising or media, particularly search impressions and clicks. This holds true for many aspects of our campaigns, but the truth is far more complex than most people believe, due to SEM’s own unique nature. In many cases, for many of our keywords, search engine inventory (how many searches occur), listings’ CTRs (define), and resulting conversion rates are as much the result of non-search on- and offline marketing as they are products of spontaneous activity engaged in by searchers uninfluenced by external factors.

What on- and offline factors have caused you to engage in searching activity? Think hard about whether your searching is spontaneous or prompted by a side effect from some part of your life. This interaction effect between advertising, marketing, PR, buzz, word of mouth, and search activity has a profound impact on campaign management best practices in paid search campaigns as well as messaging on organic landing pages.

One way to think about this interaction effect is to view search impressions and increases in CTR as a metric of your brand recall and awareness. Changing your outlook from thinking about search purely as a cause (of orders, leads, registrations, visitors, and page views) to thinking about search as an effect can be very liberating. Whatever other marketing, advertising, PR, or CRM (define) you do will create change in search activity and responsiveness.

We’ve engaged in testing and measuring the interaction effect between search and paid media with some of our larger clients (those who spend more than $1 million per month on both search and other advertising), and we’ve arrived at several hypotheses. They may not hold true in every case, but thus far the data are compelling.

Before I delve into the hypotheses, imagine traditional agencies using search as an ad success metric. If, in fact, a significant percentage of search activity is the result of external factors, then clearly search impression levels and changes in CTR are as much a measurement or metric for the effectiveness of other media as they are the foundation of an SEN campaign’s internal metric. Yahoo’s Buzz Index and custom reports from comScore, Nielsen, Hitwise, and others may be some of the best ways to judge ad effectiveness, because search behavior is such a high predictor of interest and eventual purchase.

My hypotheses:

  • Advertising interaction. Some significant percentage of search activity isn’t spontaneous at all but rather stimulated by other on- or offline advertising. When a person searches, that search may not have been fully spontaneous but was instead a result of advertising, direct marketing, PR, or word of mouth. All are forms of marketing. Search activity may be proximal to media exposure or response to the message may be significantly lagged. No one formula can be expected to work for all industry segments. Yet we can identify and track visitors based on geography more than ever. This allows us to gain an understanding not only of an SEM campaign’s effectiveness, but also the media buy’s and offline ad creative’s effectiveness.

    If post-click conversions are the only behavior you want to link to advertising, you may not need to change campaign structure. However, if ad CTR is also of interest, restructuring your Google or MSN campaign makes sense (Yahoo’s localized campaigns are quite different and probably not worth touching for the purpose of media measurement).

  • PR interaction. This hypothesis is an extension of the first and allows for the fact a significant percentage of search activity isn’t spontaneous but stimulated by PR, buzz, or news. When building a media measurement model that attempts to measure the first hypothesis, it may be prudent to also measure PR, either nationally or within geographical segments.
  • Word of mouth. They used to call it viral marketing. Now they call it word of mouth (WOM) or buzz marketing. No matter what you call it, buzz generated as prospects and customers talk about your company or product can be a powerful influence on search behavior and, therefore, on a search campaign. WOM could initially be stimulated or generated by PR or advertising or could have its cause in the product itself. Think about Google’s first four years, a period in which it grew from WOM almost exclusively. If you’re going to use search as a success metric for other forms of marketing, don’t neglect WOM’s the potential impact.
  • Search. Our own data, as well as data reported by comScore, show consumers engage in long strings of searches as they move down the purchase-decision funnel. Our hypothesis is that search impressions and engagement with a marketer’s site stimulate additional search behavior. When the advertiser’s site has been engaged with, future search interaction is positively weighted toward the brand/site with the initial engagement. This behavior corresponds with classic buy-funnel behavior.

Not many marketers are ready to use search as a success metric for all their various media and PR campaigns. Yet understanding interaction effects between media allows a good SEM agency or in-house search team to leverage all other media and marketing campaigns to improve SEM campaign efficiency and effectiveness.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.

Related reading