Domainers have long sampled domains by “tasting” them for traffic after they drop to see if they have any revenue potential before committing to a purchase. (EDIT: My friend John Andrews pointed out that domain tasting has been over for a while, but I’m sticking with the analogy:) )
The same principle applies to PPC and can represent one of the fastest and most profitable ways to scale out your campaigns.
The “Keyword Tasting” technique is one of my favorites. You grab a fairly broad type of keyword and toss it against the wall in a new adgroup to see:
a) is there any traffic there
and
b) can the keyword net ANY conversions.
We’re not worried about ROI at this point, we just want to see if there’s any juice there.
I typically let the tasting go on for a few days or enough time to capture enough data to get a look at the “actual search terms” in the Adwords Keywords tab. For most of my campaigns this ends up being a few days and 300-500 clicks.
If a few conversions roll in the broad term, have a look at the specific search terms that netted the conversions and add them as an exact or phrase match to the adgroup (or add them to a net-new adgroup if you like).
If the broadish keyword doesn’t result in any conversions after a decent amount of data, cut it. You’ve just saved yourself a ton of time going “deeper” on a keyword that most likely doesn’t have any conversion juice.
If the “throw it against the wall and see if it sticks” approach results in some conversion potential and you can see via the “See Search Terms” button which keywords and match types grabbed those conversions, you’ve got some great new adgroups to expand with the foreknowledge that they can convert.
In a future post we’ll expand on the ‘wide then deep’ approach to keyword expansion. In the meantime, why not try this out today with a new broad keyword and see if you can find any gems!
Hi Geordie,
As long as you’ve got the money to experiment with, this works great.
Now that Google Analytics has more AdWords features you don’t have to wait a few days to see your ‘matched search queries’. So, I wouldn’t recommend the set it and forget it approach if you have the time to look for negatives (might save a few bucks out of 500 clicks).
I look forward to your “wide then deep” approach to keyword expansion.
09.07.10 at 4:08 pm
Thanks for the comment Chad,
The additional time and extra clicks do give you a higher probability of getting actionable data on the conversion rates of the keywords you’re being matched to. These conversions might be missed if tests are cut short early on.
09.07.10 at 11:05 pm
Did you read Dave Naylor’s blog today?
What do you think about the (duplicitous, lying, dishonest, misleading, sleazy) inaccuracy of the Google Adwords Keyword Tool? Last month, some folks on Dave Naylor’s blog showed how inaccurate the Google keyword tool was:
http://www.davidnaylor.co.uk/why-the-google-adwords-keyword-tool-shouldnt-be-used-for-seo-a-case-study.html
I do SEO and rely exclusively on this tool to determine keyword demand. Now it seems like Google is showing more “accurate” results:
http://www.davidnaylor.co.uk/is-google-adwords-keyword-tool-now-more-accurate.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+davenaylor+%28David+Naylor%29
How has this effected you as a paid search marketer?
It reinforces what Perry Marshall says about Google… sleep with one eye open.
Raza
09.08.10 at 3:29 pm