Broad Match is Pay Per Click Fraud?
December 28, 2008
Adwords advertisers may be paranoid of pay per click fraud:
- competitors may maliciously click repeatedly on your search ads
- content publishers may participate in click fraud schemes
- "click bots" clicking on every ad on a page
photo by: Adam_T4
However, did you know that good old “broad match” keywords may actually be responsible for the biggest “click fraud”, in everyday practice? Technically, its not really fraud though, as Adwords does spell out the risks of using broad matching… “Keyword variations can include synonyms, singular/plural forms, relevant variants of your keywords, and phrases containing your keywords.”
My gripe is with the "relevant variants"... Broad matching goes too than that though, much further… and most advertisers are left clueless as to understanding the missing puzzle pieces of this irrelevant matching game, or "emptying your wallet scheme"…
Let's review some real examples...
Let’s say you are bidding on the keyword Party Costumes, in broad match of coarse. (the default keyword option)
Not only will your ad appear for relevant searches like… party costume, costume parties, ect… but your ad will also appear for “relevant variant" like...
- party ideas
- party customers (misspellings also are fair game)
- party decorations
- new years eve decorations ( its kinda related, ya?)
- and foreign languages too!
You get the idea… The Adwords Broad match algorithm considers its all fair game; anything slightly related is marketable. Its not just synonyms, it can be a very loose connections, very loose indeed. Like “shopping displays” is related to “shopping carts”?!
With over 50% of searches being unique, perhaps this is Google’s method of "capturing" all available income from their advertisers?
Unless you utilize server logs or exact queries with an analytics service, you are basically blind to the situation, basically chalking up your poor ROI to the price of playing the game.
Even if you utilize the search query report, you most commonly see these long tail queries as: “other unique queries”. Jeez, thanks alot. Give it up already. We want all the queries, including the “outliers” please. In aggregate, these outliers comprise the majority of our clicks/budgets, if you bid on broad keywords.... BE FOREWARNED!
I would actually guess that more than 80% your Adwords budget is wasted on irrelevant clicks according to this nasty broad keyword algorithm.
Its a catch 22 though, as advertisers do indeed wish to capture relevant “long tail” search, and this is most possible with broad match; however, with the current situation as it is, only very skilled and diligent adwords managers would be able to avoid this "broad keyword trap".
We implore Adwords to return to the original broad match algorithm, ie. to capture queries that are indeed relevant… singular, plurals, and true synonyms, not “related words”, like “tennis sneakers” is related to “slippers”, as they are both footware…. JEEZ!
Google, if you are listening, let’s get back to basics, ok?
Relevancy is King, not Q4 profits.





2 comments:
Yeah two ways I like for combating this are:
Aggressive use of negatives
A sort of "expanded" phrase match where you generate every variation of a phrase (so best dog food becomes "best dog food". "dog food best", "food dog best", etc.) This way you capture more traffic without trying to sell slipper buyers your tennis shoes. Nice post about it here.
Like you said mining log data/analytics is a great way to find negatives; my company (WordStream) actually has a tool that introduces query data into a dashboard from which you can assign negatives. Might be worth checking out.
Thanks!
Tom Demers
I recently came accross your blog and have been reading along. I thought I would leave my first comment. I dont know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.
Sarah
http://www.lyricsdigs.com
Post a Comment