In the UK SERPs, respondents’ accuracy in selecting the higher-ranked page was just 46%.

It was higher in the US by about 10%, but for many search terms, accuracy was frequently in the 40% range.

Since the survey included respondents with varying levels of SEO expertise, as well as laypeople, one might expect those with more knowledge to have better predictive capabilities.

But that wasn’t really the case.

For example, for UK SERPs, while SEOs with three or more years of experience did perform better than laypeople and SEOs with under three years of experience, their accuracy was still equal to that of a coin-flip.

What’s more, those who indicated that they were “sure” or “fairly sure” about their predictions were less likely to accurately predict which page outranked another for a particular search phrase by nearly 4%, and only slightly more likely to outpredict those who guessed. 

Blame the Penguin?

While these numbers will provide ammunition to SEO skeptics, and might even give companies some pause about their own SEO efforts, it’s also reasonable to look at these numbers as evidence of just how complex Google’s algorithm is to decipher today.

Google’s core algorithm now consists of over 200 unique signals that can affect rankings.

And with Penguin 4.0, Google has not only integrated one of its most important spam-detection algorithms into its core, it is refreshing Penguin’s data in real-time, “so changes will be visible much faster, typically taking effect shortly after we recrawl and reindex a page.”

That means SERPs can change very, very quickly, and frequently.

Throw in personalized search and it’s really no surprise that even experienced SEOs can’t accurately predict whether one page will outrank another.

The future of SEO

And that really isn’t what legitimate SEO is. Given the complexity of Google’s algorithm today, an SEO’s worth today comes down to two core capabilities: 

  • Being able to identify and fix problems known to be detrimental to rankings. While it might be nearly impossible to predict exactly what will produce a top ranking for any specific search query, there are numerous well-established mistakes that, when present on a site, can harm rankings.
  • Monitor and analyze data to suggest and implement improvements that might produce ranking improvements. There is no science for achieving top results, but a good SEO is capable of analyzing data and making changes over time designed to produce better results and respond to negative changes.

 In other words, companies should expect SEO to help them avoid the big no-nos and engage in educated, data-driven iterative experimentation to try to boost results.

To learn more, check out these resources: