PUBLISHED: Sep 11, 2009 3 min read

Google Broad Match Opens Wide

James Connell

James Connell

Group Director, Media + Analytics

Broad to Begin With
Google recently expanded the breadth of searches that will match to broad-matched keywords. Originally, broad match would trigger ads when a searcher typed in misspellings of an advertiser’s keywords. A little over a year ago, Google expanded broad match to include synonyms of an advertisers’ keywords. While that did enable the advertiser to have his/her ads show for search queries that he/she hadn’t thought of, it also resulted in ads showing for unrelated queries. For instance, do you think that the query: “What is event marketing” should be matched to the keyword: “hispanic advertising agencies?” Me, neither, but Google’s broad match algorithm did. This expansion turned broad match into a very powerful, but potentially dangerous tool: It could deliver new customers, but it could also eat up your budget if you weren’t careful.

Now, Even Broader!
Google has now expanded broad match so that it not only will trigger ads for queries that include synonyms of an advertiser’s keywords, but broad match will now trigger ads for queries they typed in earlier in the search session. For instance, I did a search for “send flowers nyc,” then followed it with a search for “where can I buy cheap pants.” The ad in top position? “$19.99 Flowers for Cheap.” “50% Off Flowers Today” took 3rd position.

In broad match’s defense, Google does allow advertisers to select words that will prevent their ads from showing if the words are in a search query. However, it was already nigh-impossible to predict all the synonyms Google would match ads to before this new expansion. With it, it is now completely impossible; how could an advertiser ever think to prevent an ad for flowers from showing against a query for pants?

The End Result: Lower Conversion Rates
A quick survey of our clients shows that anywhere from 5-13% of Google clicks are coming from “Session-Based Broad Match” as Google calls it. While those percentages don’t seem to be too high, they hide the drastic effect that session-based broad match can have on a campaign’s conversion rate. For instance, one of our clients, who normally enjoys a conversion rate of about 38%, had only one out of 12 session-based clicks convert over the past week. That’s only 8%!

No Opt-Out Button
In spite of the lower conversion rates, Google is not providing advertisers any way to opt out of session-based broad match, aside from doing away with broad match altogether. We are working with our partners at Google to get this changed. In the meantime, we are looking more closely at the data to gauge the change’s effect on each of our accounts, and take counter measures.