23 Mar The Holy Grail of Search (Guest Post)
In the online world, the hunt for a pure source to heal all wounds is found, like the Holy Grail, in the journey. In online search, however, the Holy Grail isn’t a cup used to consume holy water, but rather a process of testing and optimizing. Just like in Dan Brown’s The Da Vinci Code, which centers on a search for the Holy Grail, it is important to know who can be trusted and when they can be trusted.
In the paid search market, the number of experts is growing at an increasing rate, but you should be careful of who you trust and when you trust them.
Account Structure: It might seem like a pretty straightforward topic. An Adwords account is structured hierarchically starting with an account, which branches into up to 25 active campaigns, with each campaign having between 1 and 100 active ad groups. Setting up your account in a sophisticated way can save you big dollars and drive your quality scores higher. Two areas impacting quality score most heavily are:
- Relevance of the keyword to the ads in its ad group
- Relevance of the keyword and the matched ad to the search query
Most search marketers will tell you that the best way to optimize account structure is to take your best keywords and separate them into their own ad groups, so that ad copy and keywords are most relevantly matched for highest CTRs and quality scores (and by default, lower per-click costs). While I won’t say that isn’t a good approach, you need to fully understand both sides of this coin to see which is most valuable to your search program. Let me layout 3 scenarios:
Scenario 1: An account’s top keyword (performance and volume) has a quality score of 10 and is in an ad group with semi-relevant terms that have average QS of 6. The top keyword makes up 50% of the total query volume for the entire ad group. Overall, the ad group pays $0.50 / click and receives 10,000 clicks / month (total cost of $5,000) and converts those clicks at 10%. This scenario creates 1,000 conversions at a $5 cost per conversion.
Scenario 2: To improve quality score, the ad group in Scenario 1 is broken down into two ad groups. Ad group 1 is a one-word ad group with the account’s top keyword. Ad group 2 has all other keywords from scenario 1. Ad group 1 has a QS of 10, receives 5,500 clicks (10% lift from scenario 1) at $0.60 / click ($3,300) and converts at 12% (660 conversions) at a $5 cost per conversion. Ad group 2 has an average QS of 4, receives 4,500 clicks (a 10% decline) at $0.50 / click ($2,250) and converts at 7% (315 conversions) at a $7.15 cost per conversion. Overall, this scenario produces 25 fewer conversions at $5.70 cost per conversion.
Scenario 3: Scenario 3 is similar to Scenario 2, but is an example where the top keyword is a larger percent (80%) of total volume; this changes the outcome. Ad group 1 is a one-word ad group with the account’s top keyword. Ad group 2 has all other keywords from scenario 1. Ad group 1 has a QS of 10, receives 8,800 clicks (10% lift from scenario 1) at $0.50 / click ($4, 400) and converts at 12% (1056 conversions) at a $4.17 cost per conversion. Ad group 2 has anaverage QS of 3, receives 1,800 clicks (a 10% decline) at $0.55 / click ($990) and converts at 6% (108 conversions) at a $9.17 cost per conversion. Overall, this scenario produces 164 more conversions at $4.63 cost per conversion.
As you can see from the examples above, things aren’t always so cut-and-dry. There are times when the “experts” are right and there are times when going against the grain is the best idea. But don’t trust me–trust your tests.