The (Slow) Death of The Keyword

Reaching consumers based upon their keyword searches has long been the lifeblood of nearly every digital marketer. Whether it was SEO, optimizing your website to “rank” for important keywords, or bidding on critical keywords in paid search, keywords reigned supreme. The formula for successful and profitable SEM campaigns became pretty straightforward: identify which keywords help you sell more, and then dominate those keywords in your SEO and paid search campaigns.

In October 2011, Google announced “secure search” in an effort to protect user privacy for organic searches.  The impact on marketers and SEO practitioners was devastating. With this change, the organic keyword that led to a conversion or sale would more times than not appear as “not set” in Google Analytics, making organic keyword-level optimizations increasingly more difficult.

As a result of these changes, many marketers opted to take the easier path of just bidding on those essential keywords in paid search campaigns in a platform appropriately called AdWords. Then, in January 2012, Google created a new metric called Impression Share, which basically told you how many keyword searches you were missing out on based upon your current budget. This change had a massive impact on Google’s revenues, and simultaneously made paid search much more competitive and costly, which caused marketers to shift vast sums of their marketing budgets to ensure they achieved the recommended Impression Share of 90% or more. One could argue that the significant shifts in budgets were due to Google’s ability to generate more leads and sales for marketers. This may be partly true, especially when Google Analytics and AdWords, for the longest time, only measured performance based upon a last-click attribution model.

As someone who provides attribution software for the automotive industry, I can tell you firsthand that the metrics used to justify more budget for paid search were often not based upon actual sales. Instead, the budget justification was based upon keyword-level vanity metrics, like Impression Share. Of course, there are exceptions where we see automotive marketers connecting the keyword to the actual sale and optimizing accordingly.

As paid search budgets started to skyrocket due to some marketers’ unhealthy obsession with vanity metrics like Impression Share, savvy paid search professionals began to utilize Google’s search term report, which shows the exact keyword that triggered your ad, not the bidded keyword. As a result, the search term report allows marketers to identify low-cost, high-performing long-tail keywords and expose wasteful keywords. These wasteful keywords can then be added as negative keywords to ensure your ad doesn’t appear for searches for those wasteful keywords. In many cases, these negative keywords have enabled marketers to lower their paid search budgets significantly.

Last week Google announced that it will now limit search term reporting. The announcement stated, “We are updating the search terms report to only include terms that were searched by a significant number of users. As a result you may see fewer terms in your report going forward.”

Marketers are upset, and rightly so.

Ryan Everson, Corporate Digital Marketing Director at Garber Automotive Group, stated in a Linkedin post,  “Google has undoubtedly been progressively moving towards more automation and giving advertisers less manual control.

This change is slightly different from previous changes because it prevents advertisers from knowing which search terms they spent money on. It also makes it much more challenging to build out a comprehensive and ever-evolving negative keyword list.

This seems to be a way for Google to make more money under the guise of privacy.”

Google’s announcement also seems to reinforce a product roadmap designed to steer marketers to more keyword-less driven campaigns.

Why is Google doing this? I believe part of their motivation is based upon the sheer complexity of successfully managing and optimizing paid search campaigns at the keyword level, which in many cases involves managing 10s if not 100s of thousands of keywords. Suppose Google can remove the keyword complexity of paid search, by ironically making it less effective in some cases. Will less sophisticated marketers abandon keyword-only campaigns and create a paradigm shift to buy Google based solely upon broader audiences, not keywords? I believe it will. Consequently, these broader audiences will require bigger spends to effectively reach them.

In fairness to Google, judging someone’s intent, based solely upon their keyword search, is very difficult. Moreover, search has long been viewed as a lower-funnel channel, further reinforced with last-click measurement defaults in Google Analytics and Google Ads. To overcome this lower-funnel bias, Google has developed numerous audience targeting capabilities over the last 3-5 years to appeal to more upper-funnel ad dollars designed to reach consumers across all channels, not just search. In many ways, they are taking a page(s) from Facebook’s people-based marketing playbook by equipping marketers with more audience targeting options based upon additional in-market shopping signals beyond just a keyword search.

As Google evolves its targeting options away from keywords to more audiences, it appears keyword-only targeting will slowly become less and less effective in the name of consumer privacy and revenue. As a result, marketers will need to adapt accordingly by incorporating more audience targeting into their paid search campaigns. But which in-market audiences? Stay tuned. I will address that question in my next post.

RIP keyword.