Too often bid optimizations focus solely on keyword level metrics such as CPC, conversion rate and ROAS. We are all guilty of falling victim to this bid optimization pitfall. The formula is simple and straightforward: Over the past 30 days keyword “x” has seen a CPC of $1.00 and a conversion rate of 1.00%. My acceptable CPL is $50.00, therefore my bid for keyword “x” need to be decreased by 50%.
The logic makes perfect mathematical sense. And in a vacuum, this strategy will be extremely successful. Unfortunately, PPC does not operate in a vacuum and there are other factors that need to be considered in order to properly optimize a bidding strategy at the ad group, campaign, or account wide level.
The most important factor to consider here is keyword match type and how that affects which keyword in your account actually triggers an ad (thus resulting in click, CPC, conversion, etc. to be attributed to that keyword).
Consider the following set of keywords for the fictional Westborough Technical College. All three keywords are assigned the same CPC bid of $1.00:
+trade +school +courses
+trade +school +classes
+trade +school +programs
While conducting some research on potential options after graduating in June, a high school student types “local trade school programs” into Google. Intuition tells us that of the keywords listed above, +trade +school +programs will likely trigger an ad. Google, however, does not serve ads based on the most closely aligning keyword within the account, it serves ads by selecting the eligible keyword that will result in the highest CPC. They are, after all, in this to make money.
If +trade +school +programs does happen to win out in the auction and serves ads for the “local trade school programs” search, we are able to gather data and make a good valuation of that search’s value after gaining enough volume to warrant statistical significance. We come to the same conclusion as in the example outlined at the beginning of this post; this keyword’s bid needs to be decreased by 50%.
Although still eligible to serve ads for the “local trade school programs” search, you will notice that we are now bidding 50% less on this term than other keywords in the group. The kicker here is that based on Google’s match type rules, the other two keywords are eligible to serve ads for a “local trade school programs” search, too. And, given that Google will always serve ads to the eligible keyword with the highest CPC, this query will now be directed to one of the other keywords. So in essence, we have not devalued the search query itself, only changed which keyword would tally the impression. We are still paying $1.00 for a search we have already established as having a value of only $0.50.
This is why I use a method focusing on keyword intent (or likelihood of conversion) when structuring my campaigns and ad groups. But grouping similar intent based keywords into the same ad group, I can make ad group wide bid adjustments to ensure I don’t overpay for a click that I already know is not worth the squeeze. This makes keyword level bid optimizations both more powerful and more efficient.
By taking the time to structure the account properly and apply sound, well planned logic, I can analyze the value of a search on the intent level rather than the keyword level. By creating tiers of intent, I can quickly and efficiently establish that (based on the max. CPL of $50.00) tier 1 keywords (Westborough Technical College) convert at 5.00% and assign them a $2.50 bid, tier two keywords (technical college courses, technical college programs) convert at 2.50% and assign them a $1.25 bid, and tier 3 keywords (trade school courses, trade school programs) convert at 1.00%, so they get a $0.50 bid.