As I audit campaigns, I generally look for three main things: ways to reduce inefficient spend, 
ways to improve current efforts, and opportunities to grow the account. After an audit, it’s fun to talk to the client about findings. Through many of these conversations over time, I’ve come to find that there are a few misconceptions about ‘best practices’.
Let’s talk about 5 of the common recurring misconceptions that we find ourselves dispelling.
Exact Match is Always More Valuable Than Phrase or Broad
In theory, Exact match is a more valuable keyword than phrase and broad when it comes to relevancy. However, that’s generally not the case as it pertains to volume. Cost efficiency can really go either way – sometimes exact match produces a cheaper cost per acquisition and sometimes it doesn’t. We sometimes see accounts that are built out in exact match alone.
These accounts usually struggle to scale because volume is so limited in their current condition. Conversion volume is often somewhat inconsistent which prohibits the CPA from leveling out. These accounts benefit from the addition of select phrase and modified broad keywords.
With Broad, No Other Match Types Are Needed
For as many accounts with exact match only, or predominately exact match, structures we see the same on the other end of the spectrum. I can’t count the number of accounts that I’ve reviewed that are 90% broad match or more. While broad match will capture the volume, it won’t generally be the most efficient. Especially if the keyword are regular broad, not modified. A fully broad keyword strategy requires heavy lifting with negatives but that’s not generally the case.
Likewise with the first bullet, these accounts benefit from the addition of other match types to better control relevance and query mapping. (Not to mention a lot of negatives.) Broad match brings in the volume but adding negatives and additional match types can help to refine the cost per acquisition.
CTR is the Most Important Ad Metric
CTR is an important ad metric but unless clicks are the main goal, there should be other conversion metrics in consideration, as well. I like to look at conversion per impression as opposed to clicks per impression, which ensures that we didn’t only get a click but also a conversion. If you aren’t using leadgen, you can look at profit per impression. Brad Geddes has endorsed this theory for a long time, so no need to take my word for it! One caveat is that you might not want to choose an ad with a super-low CTR in case it might impact quality score. Still, though, CTR shouldn’t always be considered the most important metric.
As a secondary note, I often see ad rotations set to optimize for clicks. Not only does this rotation make it harder to test ads, it also focuses on the wrong metrics. Just because a searcher might be more likely to click doesn’t mean that they are more likely to convert.
Position Preference, Or Lack-Thereof
I often see accounts in extreme average positions – either all terms are in position 1 or most, if not all, keywords are in positions below 5-6. Your average position does matter, it has a direct impact on your impression share.
Now, do you always need to be in position 1? No, that’s more of a vanity metric than anything. In position 1, you may be spending more than you need to for the same amount of conversions that you could potentially receive in position 2 or 3.
On the flip side, while it does sometimes make sense to be in lower positions for the sake of cost efficiency, visibility is important. A lot of times when I see keywords in position 6 or worse, it’s because the advertiser feels the CPC is too high. The general response to this is that we need to focus on CPA as opposed to CPC.
Keyword position should really be decided on a keyword-to-keyword basis, based upon the value and cost efficiency of the keyword which can only be determined based upon that keywords performance.
Always Trust Automation / Never Trust Automation
This is another misconception that people seem to be polarized about. Sometimes we find people are entirely opposed to any form of automation, including things like conversion optimizer, and sometimes we find people have a “set it and forget it” mindset. There’s generally a happy medium somewhere in the middle. Scripts and bid rules (especially from advanced bid management platforms) can be valuable. They might not always work but if you don’t occasionally test, you’ll never know!
