Keyword Kevorkian: When Keywords Should Be Put Out of Their Misery

In nearly every paid search account there are keywords that are costing far more than they are returning – week after week and month after month – and yet are never paused or deleted. Why? This is not a small problem. I frequently see 20%, 50%, and even 70% of spend in PPC accounts delivering negative ROI and even ROAS on a per-keyword basis – and yet the responsible keywords aren’t killed. Why are advertisers wasting money in such a blatant fashion? How far would Google search revenues drop if all negative return keywords were paused? Paid search, and online advertising in general, is supposedly better than traditional media due to direct measurability and the relative ease with which costs and revenues can be aligned. In theory this leads to better decision making and increased efficiency. In reality this is sometimes true but often it isn’t. There are many reasons why money losing keywords are kept running:

• Branding – When doing business in a certain market segment many believe it’s important to ‘show up’ and be seen when customers search for related products and services. The hope is that this ‘branding’ has a positive impact later in the customer lifecycle or relationship.
• Ego – Following the same basic logic as ‘Branding’ except without the genuine or even delusional expectation of eventual results. Ego-driven keywords are there to make yourself or the boss feel better, or to create an impression within the market but not with customers.
• Lack of Faith In Revenue Tracking – Often it’s known or suspected that revenue numbers just aren’t right. It could be latent or offline revenue that isn’t included in the system, tagging problems on landing pages or the site, or other problems that undercut the credibility of the reporting system to the point of paralysis.
• Attribution – Last-click revenue attribution just doesn’t make any sense and yet it’s still the industry default and the only option many have. A linear or more sophisticated method of sharing revenue across multiple touch-points witin PPC or across channels isn’t yet perfect and there are many willing to endlessly debate the complex reasons why. Lacking any ability to give partial credit to keywords that don’t directly drive the sale, many assume those clicks probably deserve some credit and use that as an excuse to keep the spend going.
• Top Line Focus – For psychological reasons I can’t fully understand, many people seem to still believe in losing money on every sale and making it up in volume. Of course, the reality is their investors care about the top-line, they dream of lifetime value, and there are other unnatural influences at work too.
• Fear or Shame – It seems many paid search managers have been conditioned to feel like turning off keywords is wrong. Keyword expansion is great – more keywords! Killing them seems like defeat, an admission of incompetence, or the death of some potential (albiet unrealized) revenue opportunity. If business failure is supposed to be celebrated, the message hasn’t got to keyword managers yet.
• Nobody Noticed – I’d love to know what % of all ad groups in AdWords have not been viewed in any meaningful way by a human in weeks, months, or years. From entire accounts that are almost never worked on, to huge ones that get attention every day but still have ‘tail’ adgroups and keywords that just don’t merit much attention, there is a huge gap between needed management resources and available management resources.

I’m sure there are others reasons too. It’s time for paid search managers to begin addressing these issues. It’s time to stop wasting money on under-performing keywords. Obviously there is a unique reaction/solution for each of the rationals that are used to not kill a keyword. In the next few posts we’ll take a deeper dive in each and talk about some guidelines you can use to make faster kill decisions. What about you? How do you approach killing keywords? Let us know in the comments.