Just over a month ago, Google announced the global roll-out of an update to the AdWords algorithm, which increased the value of landing page relevancy and worth when determining Quality Score.

Google predicted that the changes would alter keyword Quality Scores and ad positions for some campaigns. However, the company claimed that most brands would not see a significant change in overall performance.

Marin Software decided to look into the matter. We undertook a stringent pruning process to filter out as many external factors as possible, focussing on accounts which had received consistent traffic through the period of study resulting in a sample population of 240 accounts. 

These accounts have received similar daily impression volumes, consistent number of active keywords and limited bid movement. We then analysed the daily impression-weighted Quality Score at the publisher account level. 

QS is not available as a reportable metric as a function of time at Google; this analysis required reports to be pulled daily to allow for a retrospective analysis.

As QS impacts AdRank directly, looking for any significant position shifts around the same period may also be a suitable proxy to impression-weighted Quality Score, provided the Search Bid (as well as account structure and several other factors) has stayed relatively constant in that time-frame.

Less than 5% of the accounts studied were observed to have a statistically relevant increase in Quality Score greater than 0.25. When taking a closer look at two of these accounts, we see the spike in Quality Score occurred between the October 2 and October 4. 

Furthermore, there was little change to Click-Through Rates during this time, which suggests that the increase in Quality Score was related to the quality of their landing page.

We also identified that approximately 6% of the sample accounts noticed a week-over-week drop in Quality Score of 0.25 or more.

Looking into four of these accounts further, we discovered that the drop in Quality Score again took place between the 2nd and 4th October, with very little change in CTR.

It therefore seems that the landing pages for these accounts were deemed by Google to be less relevant, which had a negative impact on their quality.

Overall then, our findings suggest that the new AdWords algorithm has had limited impact on Quality Scores overall, with only around 11.25% of Google Accounts seeing the algorithm change actually affecting their score by more than 0.25.

For those brands who did notice their campaigns experience a decrease in Quality Score during the 2nd to 4th October, with barely any change to CTRs, we would suggest they improve their landing pages to account for the change in Google’s algorithm.  

In addition, it's worth checking that the landing pages are appropriate and relevant to the keywords and creatives.. The update to the algorithm was naturally significant enough to warrant something of a fanfare from Google.

However, the impact does seem to have been minimal and the good news is that those that have seen a shift can certainly rectify the situation. 

Ed Stevenson

Published 24 November, 2011 by Ed Stevenson

Ed Stevenson is Managing Director (Europe) of Marin Software, a paid search technology firm, and a contributor to Econsultancy. He also writes the Big Search blog. 

20 more posts from this author

You might be interested in

Comments (2)



Nice bit of research but what happened to cpc? Quality score is relative to competitors in the auction. So if the drop was experienced by a wide variety of advertisers (especially those whose ad rank is within close proximity) then any direct consequences should be negated right?

over 6 years ago


Joe Southin, Director of Analytics at Marin Software

Hi Harvey. Indeed CPC is a key variable in this analysis - we actually ensured that average bid stayed constant throughout the analysis by filtering out accounts which had noticed significant bid movement - as such the findings already account for CPC. If all advertisers were similarly affected then the consequences would indeed be minimal - this article aims to highlight the fact the although the vast majority of accounts are unaffected, certain accounts did appear to be impacted so care should be taken to ensure landing-page best practices are followed.

over 6 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.