This week, I had planned to write an article entitled, “The Five AdWords Features You Should Be Using Daily But Probably Aren’t.” The post was going to expose some of the hidden, advanced features available in AdWords that were often overlooked by even the most senior AdWords pros.
Don’t miss the forest for… well, you know.
That was the plan… until I did an audit of a big AdWords account. You see, as I looked through this $10M+ annual spend account, I was pleasantly surprised to see many of the advanced features I was going to advocate for in the article were being used — but horrified to see that most of this great functionality was being nullified by terrible account structure setup.
Functionality Gone Awry
First, this particular account had a very comprehensive keyword list — over the last two months, the account had gotten clicks on more than 240,000 queries!
The problem, however, was that many of these “unique” queries were actually the same query, served over and over again in different ad groups and different match types. Indeed, I counted 90 separate queries that each showed up more than ten times throughout the account and five queries that showed up more than 40 times each.
The next problem was matching keywords to queries. The account was heavily brand-focused — 97% of the revenue in the account came from brand terms. And yet, I found numerous instances where a brand keyword was getting matched to a non-brand query and a non-brand keyword was getting matched to a brand query.
Lastly, the account had lots and lots of ad groups, but all the ad groups had anywhere from 25 to 100 keywords in each one; and frequently, these keywords were a mishmash of different match types, product keywords, brand keywords, and generic keywords.
All of this poor structure destroyed the significant work the SEM team was putting into the account working on granular site extensions, keyword-level tracking URLs, and device-specific bidding. It was a case of an SEM team missing the forest for the trees.
The Solution: Return To Account Structure Basics!
The solution to getting this account back on the right track wasn’t more advanced functionality but rather a return to the basics — in this case, a serious account structure redesign.
For example, any time an account has the same query showing up dozens of times throughout different ad groups, the SEM team should go in and evaluate the data and decide to put that query in one ad group and negative match it from the rest of the account.
There is but one optimal ad text, landing page and bid per query. That doesn’t mean you can’t test out different ad text and landing pages for your queries, but when a query shows up all over your account, you are not testing — you are letting Google match you on random queries with disparate ad text and landing pages.
Imagine that your keyword is [blue widget] — your ad text should mention [blue widget], and your landing page should take consumers to a blue widget landing page. Having this query matched against [widgets] in one ad group, [acme widget company] (your brand keyword) in another, and [blue items] in a third will lead to sub-optimal performance — period.
Having your brand terms matched on non-brand queries and vice-versa leads to the same problem. Brand keywords are special — they indicate an incredibly high level of consumer intent. Brand keywords don’t need to do a hard sell on your unique differentiators like a non-brand keyword does; in many instances, they are navigational terms designed to just get someone to the right page to purchase something.
So, having a brand term like [acme widgets] get matched on a query like [best blue widget] is a big problem, simply because your ad text and landing page assume that the user is already familiar with your product.
Good account structure will ensure that your brand terms will never show up on a non-brand query, and vice-versa. This can be achieved by adding brand negatives to all of your non-brand ad groups and non-brand negatives to your brand ad groups.
When we create ad groups at 3Q Digital, we put our best queries into Single Keyword Ad Groups, or SKAGs (technically, they should be SQAGs — Single Query Ad Groups, but that doesn’t sound as good). An ad group with 25 or 50 keywords in it is only useful as a testing ground to help you discover which queries are winners (and should go into SKAGs) and which are losers (and should be negative matched across all campaigns).
An ad group with an array of marginally related keywords (think [blue widget], [large blue widget], [widget store], etc.) does not allow you to create ad text and landing pages related to specific user intent. As a result, you’ll find some queries that do quite well in the ad group, and others that, while they might perform well in a targeted ad group, perform poorly in an untargeted ad group.
Advanced AdWords features can work amazingly, and trying out the AdWords Product Team’s latest invention is always fun. No advanced feature, however, can save you from bad account structure. In this age of dynamic site extensions, RLSA, and complex day-parting, AdWords success is still very dependent on proper organization of queries, landing pages, bids, and ad text. Until you get that stuff right, everything else is a distraction!
Image free & accessible to all from Morgue Files
Contributing authors are invited to create content for Search Engine Land and are chosen for their expertise and contribution to the search community. Our contributors work under the oversight of the editorial staff and contributions are checked for quality and relevance to our readers. The opinions they express are their own.
