Behind-the-Scenes at Google AdWords 2.6 -- Why are some Google advertisers complaining?

By Publisher Anne Holland

Throughout last year Google fiddled with the programming behind AdWords in its never-ending quest both to offer surfers more relevant listings and to make more money for shareholders.

AdWords expert Andrew Goodman (see our review of his latest book plus a giveaway hotlink below) calls the current situation AdWords version 2.6.

I've gotten loads of upset letters from Sherpa readers on this topic -- mainly folks frustrated (and sometimes outraged) because Google's asking them to pay far more per click than they were expecting to. So I called up Andrew to ask for an explanation.

Andrew says, "I'm seeing a number of 10-20 cent charges for clicks I incurred on terms where I was the only advertiser. In essence this means there are no consistent minimums. There are now times where you'll pay .20 for what used to be a .05 click." (And, in fact, I've met AdWords clients who have had to pay a heck of a lot more in the same situation recently.)

Plus, it seems that low-budget marketers tend to have their campaigns deactivated by Google more frequently these days. Andrew says, "You're punished for not bidding high enough, for trying to bottom fish for cheap clicks. Under the new system there is less cheap inventory available, even for clever marketers."

This in turn hurts marketers who use averaged cost per click data to convince their bosses to invest in more expensive terms. You could afford to take a hit on ROI for a few high profile search terms because of the bargain basement terms mingling in your spending mix. That's not as easy anymore.

Andrew explained to me that the way AdWords 2.6 decides where your ad will be ranked (and what you'll pay for a click) now seems to be determined by the following equation:

Your chosen maximum bid cost per click (CPC) multiplied by Your ad's "quality score"

According to Google, a quality score is "determined by the keyword's clickthrough rate, the relevance of the ad text, historical keyword performance, and other relevancy factors." Which to my own mind is Google-ese for "a bunch of stuff we won't explain the detail so marketers can't jerryrig the system and so we can change the specific formula whenever we decide to."

Like most heavy search marketers Andrew has gained some behind-the-scenes insights while trying to reverse-engineer the current formula. His tips:

- Your landing page's relevancy to the search term is probably affecting your cost and rank. If you're using a generic landing page for a wide array of campaigns, your ad rank may fall and your cost per click may rise. Your campaign could even be disabled.

- In fact if your landing page contains anything that would hurt you if it were being used to attract organic listings (ie. bad SEO) your AdWords ads could be penalized.

- The visible URL you put in your ad is probably considered part of the ad text by Google, so relevant words in that will help you. (I've certainly seen plenty of data showing the right wording in a click link can help clickthrough rates.)

- If you've been running ads for a particular search term for a year or more, Google may be less likely to penalize your rank for temporary downward blips in clickthrough performance.

- Your overall account lifetime might matter as well. If you keep switching AdWords accounts (or perhaps switching SEM agencies if they use their accounts for your campaigns) then you lose your history each time and risk your ad rankings and costs. - If you're using somebody else's trademark terms in your ad, your rank could fall, so test with and without trademarks.

- Test dayparting to turn off your ads during times of day (or days of week) when your clicks are historically lower.

- Try using Google's dynamic keyword insertion tool to get higher rankings.

Last but not least Andrew says when you launch new campaigns, you might try bidding more aggressively than you plan to for the long term just to get the highest possible ranking and "demonstrate" to AdWords that your ad can get great clickthroughs when it's in a good position. You're paying to establish a "this is a good ad" history with the program, which perhaps will give you a leg up once you take your bid back down in future.

Andrew notes, "It's a bad time to be a new advertiser. There are times you just set a budget and burn it in order to get where you need to be." He suggests that you "keep a lower ROI but high clickthrough rate ad in rotation with one-two lower clickthrough rate, higher ROI ad just to keep the quality-score wolves at bay."

In the meantime, Google's competitors are also studying the 2.6 system and deciding how much of it to copy in their own programs. If you're a major PPC spender, now is the time to get on the phone with an account rep at Yahoo/Overture and the new MSN Search and see if you can influence their development.

You know, sometimes I'm awfully glad that I'm just leading a research team instead of having to actually conduct online ad campaigns myself anymore. Today is definitely one of those times…

Here's that link I promised to our review and giveaway contest for Andrew Goodman's new book: http://www.marketingsherpa.com/sample.cfm?contentID=3117 (Giveaway contest ends 1/23)

Sponsor: Eyetracking & Conversion Data on Search Marketing ~~~~~~~~~~~~~~~~~~~~~~~
Sherpa's 2006 Search Benchmark Guide has 210-pages of stats: - Eyetracking lab test color charts - Trends in CPC, CTR, and conversion rates - Do agencies get better results than in-house marketers? - SEO (Optimization) vs PPC campaign data

+ 3,271 marketers' real-life 2005 search campaign data: http://www.sherpastore.com/Search-Marketing-Benchmarks-2006-SEO-PPC.html or call 877-895-1717
~~~~~~~~~~~~~~~~~~~~~~~