Have you ever sat down to run through your Adwords or other Paid Search account only to find yourself confused and disoriented? Do you feel like you have way too many Campaigns, Ad Groups and Keywords that your head starts to spin even at the thought of working on your account? If so, your Campaigns may be larger than they really need to be. You may be fighting a battle with yourself where you are over-thinking your Paid Search strategy altogether. Then again, who am I to say that? I’m the guy that is constantly driving my department to expand their Ad Groups into more relevant groups. We all know by now that relevancy is what drives our accounts…
- Without relevancy we don’t get better Quality Scores…
- …without better Quality Scores we don’t get improved Ad Rank
- …without improved Ad Rank we don’t get prime ad placement
- …without prime ad placement we get lower Click Through Rates
- …with lower Click Through Rates we get less traffic
- …with less traffic we are forced to increase our Cost per Clicks across the board
- …with increased Cost per Clicks we get higher ad placement, then more clicks, then more Conversions, but all at a higher Cost per Acquisition….
…so again, I think we can all agree that relevancy is what drives our paid search accounts. That said, is there such a thing as “over optimizing” your paid search accounts? The answer is yes. Here are a couple ways that you can over optimize your PPC accounts:
- Not clearly dispersing your keywords, products and/or ads well enough will lead to oversized campaigns and ad groups.
- Blindly stacking campaign level and other bidding adjustments.
To some, the first option may sound like the opposite of optimizing. Generally when people think of optimizing something they are thinking of ways to simplify them, to ty and reduce clutter. To most that aren’t familiar with paid search and optimizing techniques, generally bundling together keywords into a single Ad Group usually seems like the proper way of optimizing. This couldn’t be further from the truth. Overcrowded Ad Groups can lead to a number of different issues that can lead to a multitude of independent problems. The first, and most important in my opinion, is keyword relevance. For those that have read more than one of my blogs you will recognize that this is one of the main things I constantly preach about (as you can see from earlier in this article). Low keyword relevance is the snowflake that gets the ball rolling to the “series” of issues that I have presented earlier. Another issue equally important is that keyword overcrowding usually leads to a certain level of “low search volume” terms in your Ad Groups. Google automatically will deactivate Low Search Volume terms until such time that they deem those keywords as statistically significant. Meaning, you will have a number of terms cluttering up your ad group that have zero significance. Not to mention the fact that they are more than likely to be irrelevant to the higher traffic terms in your ad groups which will house lower Quality Scores and ultimately eat away at any chance of positive historical data. Here is the game plan that you should be working with:
- Assemble extremely relevant keywords to one ad group using Phrase Match, Exact Match, and Broad Match Modified terms. For example, use the following terms in just one ad group:
- [red shoes], “red shoes”, +red +shoes
- [red shoes for sale], “red shoes for sale”, +red +shoes +for +sale
- [red shoes sale], “red shoes sale”, +red +shoes +sale
- [red shoes near me], “red shoes near me”, +red +shoes +near +me
…anything having to do with another color, style, and brand you will want to have in a different ad group and possible even a different campaign should you choose. This step along with handle any long tail keywords and unique or unknown search queries that may present themselves over the course of time
- Monitor your Search Query Reports regularly to ensure two things:
- First, you need to monitor your SQR reports to make sure that you are adding any new negative keywords that will need to be excluded from your campaign or ad group.
- Second, you will want to find these “long tail keywords and unique or unknown search queries” that I mentioned just a moment ago so you can determine if they have any significant traffic.
- Once you find these terms in your SQR reports you will want to add them to their own ad group following the previous steps.
- This will lead to increased Quality Scores and,
- Better targeted traffic…
- …that you can send to a more precisely relevant landing page.
Next on the list is “blindly stacking campaign level and other bidding adjustments”. You will need to ask yourself if your bid adjustments are working together rather than working against one another. People way too often blindly set different bid adjustments in their accounts without analyzing why they need to be made in the first place, and without a plan of how it is that they are going to track the results once they do set them. First of all, let’s take a closer look at the different kinds of bus adjustments that can be made:
- Max CPC (Ad Group level)
- Max CPC (Keyword level)
- Location bid adjustment
- Device bid adjustment
- Time bid adjustment
- Audience bid adjustment
- Gender bid adjustment
- Parental bid adjustment
- Age bid adjustment
…now, some of these bid adjustments may not be relevant to every individuals campaign as some may only be relevant to remarketing or display campaigns, and some may only be relevant to search campaigns. That said, the first 5 generally relate to all, and are the ones that I worry about the most and try to keep on my radar.
The most important thing to remember when setting any type of bid adjustment is to make sure that you understand “why” you are making it, and then to have a plan to track the results. You should really never apply multiple bid adjustments to anything before starting out with just one. If you start with your keyword level bid adjustments and gather useful data there, you can now try stacking another type of bid adjustment on top of that. The trick here is to never add more than one variable to the equation at a time. The more variables that you add the greater the chance that you will never know what variable actually caused the resulting change whether that is good or bad.