Case Study: Overture vs AdWords Pay-Per-Click in 2002
How a small software company split its budget between Overture and Google's new quality-based AdWords auction, and what the split taught them.
Consider a small company selling a project management tool to other businesses, the kind of product with a long sales cycle and a modest but real ad budget. By mid-2002, the company has run Overture’s keyword auction, the same pay-per-click model GoTo.com pioneered back in 1998, for over a year with steady, if unremarkable, results. This spring, the team decides to test Google’s AdWords program against it.
Why the test mattered
Google’s AdWords launched in October 2000 as a self-serve CPM program, but in February of this year Google rebuilt it around a different idea: advertisers now bid per click, as they do with Overture, but ad position is no longer decided by bid amount alone. Google factors in the ad’s click-through rate too, so a highly relevant, well-written ad can outrank a higher bidder if it performs better with searchers. That’s a meaningfully different mechanic than Overture’s straightforward highest-bid-wins model, and the team wanted to know which produced better results for their specific keywords.
Running the split
The company kept its existing Overture campaign untouched and built a parallel one in AdWords using the same core keyword list, project management software, task tracking for teams, small business scheduling tool, along with matching landing pages for each. Same product, same offer, same budget ceiling, different auction mechanics.
What showed up in the numbers
On Overture, results tracked closely with bid amount, as expected: pay more, rank higher, get more clicks, in a fairly linear relationship. On AdWords, the team noticed something different after a few weeks. Ads with tightly written copy that matched the exact search term closely were earning strong positions at a lower cost per click than equivalent bids on Overture, because their click-through rate was pulling their effective rank up without requiring the highest bid in the auction.
The practical effect was that AdWords rewarded good ad writing in a way Overture’s model didn’t. A generic ad on Overture could still win a top spot by simply outbidding everyone else. The same generic ad on AdWords, with a weak click-through rate, got pushed down the results even at a competitive bid, while a sharper, more specific ad from the same advertiser climbed past it at a lower cost.
The result
Over two months, the AdWords campaign delivered a noticeably lower cost per lead than the equivalent Overture spend, once the team adjusted their ad copy to lean into what the quality-based ranking rewarded: specific, relevant language tied directly to the search term rather than generic product claims. Overture still performed reliably and remained useful, particularly for terms where the company was willing to simply outbid competitors for guaranteed top placement.
What the team took away
The lesson wasn’t that one platform beat the other outright. It was that the two auctions reward different behavior, and a campaign built for one doesn’t automatically perform well on the other without adjustment. Bidding well on Overture is mostly about knowing what a click is worth and being disciplined about the ceiling. Bidding well on AdWords means that discipline still matters, but so does writing an ad specific and relevant enough to earn a better position than money alone would buy.
More on Paid Advertising, or see everything published in 2002.
