The Current State Of PPC Keyword Match Types
I’ll start with a big thank you! It’s been a busy few months with advertisers and Google forcefully exchanging opinions back and forth on topics such as ad rotation, features needed/wanted in AdWords, and changes to keyword match types. In my column on the Google ad rotation change last month, I expressed my strong concerns about […]
I’ll start with a big thank you!
It’s been a busy few months with advertisers and Google forcefully exchanging opinions back and forth on topics such as ad rotation, features needed/wanted in AdWords, and changes to keyword match types.
In my column on the Google ad rotation change last month, I expressed my strong concerns about the latest changes to ad rotation (as did many other paid search advertisers and bloggers) and was very pleased that Google’s VP of Product Management, Nick Fox, announced that Google had reconsidered their actions on ad rotation behavior.
First, they expanded the even rotation period from 30 to 90 days, which seems like a more reasonable time period even for low data ad groups.
Second, they made the auto-optimize ad rotation an optional setting – but only if you specifically request it. If you haven’t already done it, take a minute now to keep your ad rotation options open. We have done this for all of our client accounts and encourage you to do the same thing.
So, before I move on to today’s topic, I want to acknowledge the engineers at Google, who responded swiftly and diplomatically to its advertiser concerns.
Thank you, Google AdWords Engineers!
Match Types & Normalization – Where Are We Now?
Both Google and Microsoft have made significant recent changes to keyword matching options and the underlying normalization logic, so I thought it would be a good idea to take a step back and summarize where I think we are now, as of June, 2012.
The most significant change from Google was the addition of what I call ‘fuzzy’ matching for phrase and exact match types, which enables matching to singular or plural queries, close-stemmings and misspellings. Sort of like broad match modifier, but without having to add + signs, I suppose.
We don’t know precisely how Google decides what qualifies as a close variant though I’d be willing to bet that it gets fuzzier over time.
Meanwhile, Microsoft has made a virtual boatload of keyword-related upgrades to the adCenter on its steady march towards full functional parity with AdWords. These changes include completely revamped negative keywords, addition of broad match modifier, and fuzzy phrase and exact match.
adCenter Negative Match
Microsoft abandoned the much pilloried cascading negatives implementation back in November, 2011. Now, negatives pretty much work the same way on adCenter as they do on AdWords, except that there is no broad match negative on adCenter.
One of the best parts of Microsoft’s newer implementation is the negative conflicts report. Not only does it show where you are inadvertently blocking traffic on adCenter, but if you have just imported your campaigns from AdWords, it will also show you what negatives are blocking AdWords traffic, too.
adCenter Broad Match Modifier
This implementation is pretty much a functional equivalent of the AdWords broad match modifier implementation.
adCenter Fuzzy Phrase & Exact Match
Microsoft actually beat Google to the punch on the rollout of ‘fuzzy’ phrase and exact match. However, adCenter’s phrase and exact match is less fuzzy than AdWords, because it only matches to singular or plural versions of your keywords, not ‘close variants.’ I sure hope they keep it that way.
One of the good parts of the fuzzy match implementations from both Google and Microsoft is that they still honor negative match logic, so you can still use advanced techniques for silo-ing keyword traffic with negative match logic.
This is important for managing head terms which have significant performance differences between singular and plural keywords, and also helps to ensure proper display of DKI ads. We’ll talk more about DKI ads across both platforms in this column next month.
For example, if you have already segmented your singular and plural keyword exact match ad groups, you can prevent the fuzzy match from poaching from either of the groups by setting your negative match types like this:
This technique will work automatically on adCenter, but to keep it working in AdWords, you’ll need to explicitly modify your campaigns settings to opt out of the fuzzy phrase and exact match as shown here:
Match Type Cheat Sheet
I am a big fan of cheat sheets and summary charts, and since so much has changed with keyword matching on Google and Bing in the last few months, I figured I’d make one that captures the essentials of what’s different between the two networks.
I am sure I’ll have to update it again soon, but here is my Match Type Cheat Sheet, June 2012 Edition:
As you can see, Microsoft has come a long way towards achieving functional parity in keywords. One area where differences remain is in keyword normalization for phrase and exact match.
You have probably noticed the normalization when you’ve ported AdWords keyword lists into adCenter, when you see keywords not imported because they are duplicates. That’s where the normalization comes in, and much of it happens on phrase and exact match.
AdCenter removes extraneous words such as “a,about, an,and, at, by, for, from, how, in, is, of, on, the, to, what, with” from keywords. This is a pretty significant difference, since it means that there are vernacular phrases that work well on AdWords but will not work on AdCenter, given current normalization protocols.
For more on how adCenter currently normalizes keywords, take a look at this informative blog post from Microsoft’s Tina Kelleher.
Hope you’ve gotten some good information from this month’s column. Please do feel free to leave comments or ask questions, below. Next month, we’ll take a deep dive into dynamic keyword insertion (DKI) ad implementations between Google AdWords and Microsoft adCenter.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.