Expanded Match Woes Continue
The problems with broad and expanded match in Google adwords aren’t new but I’ve seen more discussion on it recently and I’ve also come across more examples in my own campaigns which I consider to be stretching the boundaries even more.
This article was sphunn (sphinned?, submitted to sphinn?) and generated some discussion on Rose’s blog and in sphinn itself.
It primarily talks about the problems with overlapping keywords in different ad groups. Sometimes broad match keywords can result in situations where you have carefully crafted a very targeted ad for a specific product and then Google displays an ad from another ad group unexpectedly (from your perspective). It typically occurs if you have an ad group with a general keyword like watches and another ad group with narrower keywords such as mens sports watches. If someone searches for mens sports watches then, in theory, either one of the keywords can trigger an ad as they both fit the criteria for broad matching with the search query. This is what I mean by overlapping keywords. From your perspective as an advertiser you want the ad targeted to mens sports watches but it’s not always the case. If the bid for the general keyword is higher this can be the one that is picked over the more relevant one.
It gets more complicated with expanded match. Ads can be triggered when someone searches for something related to your keywords. So, Google might consider timex watches to be relevant to mens sports watches and show your ad.
The result is more traffic, which can be good, and this is how it is justified by Google adwords. The big downside is ads showing for queries that really have no relevance to you. This lowers the click through rate and generates unwanted clicks.
The fixes are:
-
Don’t use broad match, stick to exact and phrase and the problem goes away - or
-
Include negatives to fix both the overlapping keywords problem and expanded match.
That’s all very well but:
-
eliminating broad matches will also reduce relevant traffic
-
both approaches take a lot of work to find exact matches or to find negatives
-
it’s an ongoing process to uncover all the negatives that are required. The example I had just recently was expanded match including a list of competitors business names. They showed up in the search query performance report.
Ideally I’d like the option to exclude expanded match. I can deal with overlapping keywords but expanded match is harder to control. I don’t see this being offered any time soon but who knows if the issues get highlighted often enough.
Related Posts:
Google Adwords Expanded Broad Match
Posted in Google AdWords Advanced, Keywords, uncategorised
August 24th, 2007 at 10:31 am
Hi Christine,
I ran into this by sheer chance a while back - I was managing an account that sold leather beds, and found that it appeared under a search for upholstered beds. And right below it were three other adverts for leather beds (one’s still there 6 months later!)
Google’s new(ish) tool that allows you to see the actual search term has helped to a certain extent, but at the moment, it only reports on search terms with clicks against them.
Another account that I manage, for a hotel, has a broad-match keyword generating thousands of impressions, but only a handful of clicks per day. I have no idea which keywords it’s matching to, but they don’t appear on the report, so I can’t negative-match them. And I don’t want to remove it, since it’s generating bookings at only £3 each.
If Google could just extend their keyword report to show searches with no clicks on your adverts, we’d at least have the tools to deal with the problem (in fact, we could actually make it do what it’s supposed to), but without that, I’m very reluctant to trust broad match.
I was talking today to someone on the Digital Point forums, and they’d seen their impressions increase dramatically overnight on a broad-match keyword, and they didn’t know why. I’d guess that as your Quality Score improves, it extends your Expanded Matching - if so, then no campaign is safe!
October 5th, 2007 at 6:29 am
While deleting broad match keywords should reduce impressions, clicks and sales substantially would you believe that is NOT what happens?
I’ve always used broad match keywords because originally they worked well and were responsible for right at half of all sales and revenue. (I had them tagged separately in pre-Google Analytics days.)
Since I knew that I was hesitant to delete them but finally decided to start testing it. What actually happens is surprising. I just posted the results including Account Snapshot graphs at http://www.ppcthink.com/2007/10/05/deleting-broad-match-keywords-increases-roi/
October 5th, 2007 at 7:28 pm
@custardmite: I’m late in saying this (sorry) but thanks for the detailed comment and it’s interesting to hear that you’ve had the same experience
@ Rose: Interesting! I responded on your blog post.