Search Engine Optimisation (SEO) may not be a science, but as a relatively new marketing activity, it seems to have developed its own unique language and rules. To help demystify what really matters if you want to climb the rankings in online searches, latest research from SEMrush, international specialists in competitive research for online marketing, provides new insights into search rankings following the analysis of 60,000 keywords from SEMrush’s worldwide base.

The good news is that PR really makes a difference. Describing the key finding of the study, Olga Androenko, head of global marketing at SEMrush, says: “This new study shows that the direct traffic to websites is the most influential search-ranking factor. Direct traffic is a by product of brand authority, which proves that PR and brand management are important factors in SEO these days”.

Ranking factors

Backlinks matter

Androenko says that backlink factors are also extremely important for search rankings. “Except for factors related to anchor texts, they ranked equal fourth in terms of importance. Additionally, it was found that backlink factors impact one another and that positions are unlikely to change if the values for one factor are boosted and the others ignored.

“The research also found that branded keywords will always return more relevant, but less popular, websites on the first position.”

Sometimes SEO can be confusing, but there are two simple rules for leading in searches. First, you need to have a strong brand that communicates authentically. And second, quality content matters. Two reasons why depending on PR expertise is vital if brands want to lead online.

Methodology

The second edition 2017 Ranking Factors Study analysed 60,000 keywords from SEMrush’s worldwide base, focusing on the top 17 factors using a machine-learning algorithm, the Random Forest algorithm – one of the most popular machine learning algorithms. The initial dataset used for the study consists of highly changeable values. For this reason, SEMrush didn’t use correlation analysis, which gives reliable results only when examining the relationship between two variables. For example, the impact of the number of backlinks on a search engine results page position, not multiple.