The company I work for just released a study on SERP (search engine results pages) click rates, and found that the sites surveyed received more than 95% of all their non-branded natural search traffic from page-one results pages across all three major engines. The data included 8.9 million queries sampled over nine months, representing 10 enterprise-level Web sites in many different diverse verticals.
If you are running a natural search campaign, there are several key findings here that can help put various aspects of your natural optimization program into perspective.
In this survey, only the click and traffic percentages for pages one, two, and three were scored. When looking at all three engines, what is immediately interesting is that SERP click rates were within a razor-thin margin from engine to engine. Google provided the highest average, with 95.8% coming from page one, compared to Yahoo at 95.2%, and Bing at an average of 95% (total average of 95.3%).
It's worth noting that these figures are more than 5% higher than what was observed in the 2006 AOL search data leak, and much higher than other previous research that put page-one click rates in the 80% range. Out of the almost nine million searches surveyed, queries break out across three major engines as follows:
-- More than 8.5 million non-branded natural clicks came from page one across all three engines
-- 232,000 clicks came from page two
-- 180,000 came from page three
The implications from the data are apparent: Most of your natural non-branded generic traffic will come from page one, so optimization priority should continue to focus on gaining page-one visibility.
The data also yielded some interesting click trends in the vertical area as well. Perhaps the point that stands out most is that travel sites had the highest percentage of non-branded page-one traffic (at more than 97%). Businesses that focused primarily on health and insurance had the lowest percentage of first-page traffic, coming in at 93%.
Overall search share for non-brand terms
One other interesting view of the data was in the overall share of natural generic terms that were referred by engine. In terms of overall non-brand traffic volume, Google led the way with 74.3 %, meaning that Google referred 6.7 million of the total 8.9 million queries surveyed. Yahoo and Bing were neck-and-neck at 13% of each of the remaining referred traffic. Some verticals showed varying performance in terms of share drawn from each engine. One retail survey that skewed toward people in their 20s captured more than 84% of their non-branded natural traffic from Google.
So if page one is so important, then why bother with measuring pages two and three?
If your page two and three rankings are generating existing traffic, or show the promise of traffic based on search history for those terms, then they are prime candidates for additional optimization focus in order to push them to page one. It is also important not to get too hung up on position data, as your analysis should also focus on optimization of terms based on business goals, traffic volume potential, and competition, among other factors.
Overall, this data reinforces what experienced enterprise search marketers know and practice: If you want to capture the attention of a searcher in natural search, it's page one or bust.