Conventional wisdom with most of the PPC gurus and even the major search engines is to delete or pause keywords that are not performing. Typically, I will pause keywords, adgroups, or campaigns that are not generating conversions.
The other day I discovered something interesting that yielded the same result in both Google and Yahoo.
After scanning my Google and Yahoo accounts for keywords with zero to 99,999 impressions, zero assists, and zero conversions over the previous 30 day period, I paused those keywords which resulted in approximately 50,000 keywords getting paused.
In theory, pausing or deleting keywords that are not performing at all should not have any negative affect on your campaigns, budget, or conversions.
After pausing non performing keywords, both my Google Adwords and Yahoo PPC accounts started to perform oddly. My conversions dropped and my CPA average went up by $1.50. After discovering the odd performance with both accounts I started un-pausing some of my adgroups and keywords.
This tells me there is some type of connection between account performace, impressions, and possible keyword counts.
Again, conventional wisdom is to apply the 80/20 rule. I'm going to do some more experimenting with pausing / un-pausing keywords that don't perform and try to measure the effect it has on account performance.
Are you a PPC guru? Have you encountered something similar? I currently manage about 1,000,000 keywords and would appreciate any feedback you might have for our visitors.