Among these articles, each penned by knowledgeable marketers and SEOs (or at least, knowledgeable enough to get their own article ranking on the first page for that query) there will be a variety of different opinions about which ranking factors are most important in 2019 and how heavily they are weighted.

This is because Google very rarely goes on record to state exactly which factors its search engine pays most attention to in order to prevent users from gaming its system (and, to be fair, because the answer is by all accounts too complex to be boiled down to a simple ranked list). But gaming the search engine system is what SEO is about, and as long as Google is all-important for search, SEOs will debate, theorise and pontificate about ranking factors.

Individual marketers, agencies and consultants might have different views on what SEOs should prioritise and to what extent, but there is usually some kind of industry consensus about how SEO “works” – otherwise there could never be such a thing as SEO best practices. This consensus tends to evolve over time (as Google’s algorithm evolves, and the internet changes), and taking the pulse of the industry in a particular year can provide an interesting insight into how the experts approach SEO and where they think it is going.

SEO guru Rand Fishkin (formerly of Moz, and now of SparkToro) has set out to do just that with a recently-released survey of 1,584 professionals in the field of SEO on what they believe to be the most important Google ranking factors. The survey builds on similar research that Rand carried out at Moz with the Moz Ranking Factors survey, published every two years from 2005 to 2015. The revived survey, now under the SparkToro banner, will be carried out on an annual basis from now on.

So, what do more than 1,500 SEO experts believe are the most important factors for ranking on Google, and how unified are they in holding these opinions? Let’s take a look at the survey’s findings.

The most and least important Google ranking factors, according to SEOs

Participants in the ranking factors survey were asked to rate 26 factors on a scale of 0 to 10 according to how much weight they believe each one receives in Google’s organic ranking systems (with 0 meaning “Not used” and 10 meaning “Very heavily weighted”).

According to the SEOs surveyed, the following five ranking factors are the most important for SEO, with an average rating ranging between 8.52 and 7.26:

  1. Relevance of overall page content – 8.52
  2. Quality of linking sites & pages – 7.87
  3. Use of query-relevant words & phrases – 7.5
  4. Domain’s perceived expertise, authority & trust – 7.48
  5. Mobile friendliness of the UI/UX – 7.26

The following five ranking factors are believed by participants to be the least important to SEO, with average ratings of 4.19 (which is still a fair way from 0) to 5.16:

  1. Keyword in the domain name – 4.19
  2. Use of external links on the page – 4.66
  3. Use of Google AMP – 4.76
  4. Age of the website – 5
  5. “Unlinked mentions” of the site/brand – 5.16

To contextualise this, it’s also worth noting that SparkToro asked the participating SEOs how consistently they believe Google’s algorithm weights ranking inputs across queries. In other words, do the same ranking factors apply uniformly to every single query, with the same relative importance?

The answer, in the eyes of SEOs, is overwhelmingly “no”: 66.3% of SEOs surveyed said that they believe “The weight of ranking factors varies widely depending on the query”. A further 26.4% believed there is some variation, and that ranking factors are “consistently weighted across most queries with a few exceptions”. Only 7.36% of SEOs believed that ranking factors are “consistently weighted across all queries”.

This means that although SEOs were willing to assign a relative overall importance to each ranking factor on the list, they might not necessarily believe that the same ordering applies to every single query.

Where SEOs differ about what to prioritise

A single average rating assigned to each ranking factor can only say so much about how important SEOs believe it to be: it doesn’t tell us how widely SEOs’ opinions varied around that average. Was website age rated a five because all SEOs believed it is of moderate importance to Google’s algorithm, or because half the respondents believed it is of no importance whatsoever, and the other half believed it is all-important?

To shed light on this, SparkToro used a standard deviation to rate ranking factors by relative disagreement, revealing which ranking factors SEOs are united over, versus which ranking factors they are divided over. The list shows that SEOs are generally united around the ranking factors that they consider to be most important, with almost all of the top five ranking factors achieving a high level of consensus.

The main exception to this rule was mobile friendliness, which ranked fifth-highest in terms of importance but 11th in terms of consensus, with a standard deviation from the mean of 2.07 – showing that even in 2019, SEOs can’t agree on how important mobile is to SEO. It would be interesting to be able to break this result down further in terms of demographics, in order to see how highly SEOs from different industries and localities rate mobile friendliness. (Rand acknowledges that a failure to ask about geography was one of the “biggest misses” of the survey this year). Rand himself notes that he was surprised how highly mobile friendliness ranked, as he believes that it is a “relatively small” direct ranking factor, but may “nudge” things like link earning and engagement, which are more important.

While mobile-friendliness made it into the top five most important Google ranking factors, SEOs are still split on how much it matters

 

Much less surprising was the level of division around Google’s AMP (Accelerated Mobile Pages), which ranked second only to keywords appearing in the domain name for disagreement among SEOs, with a standard deviation from the mean of 2.65. This accurately reflects the level of division among the SEO industry when it comes to AMP, with many SEOs believing AMP – as a highly mobile-optimised, Google-created web experience – is extremely important to search optimisation, and just as many believing that it should be completely ignored.

Rand noted that “One could argue that Google AMP is an “all or nothing” factor – in the AMP box on mobile, it’s essential to even being considered, and in all other cases, it makes little difference”, potentially also contributing to the high level of variation in opinions on AMP.

Links and keywords: How much do they matter?

Some of the most interesting and surprising findings of the research revolved around SEOs’ views regarding the importance of two “classic” SEO components of SEO: links and keywords.

Obvious keyword-targeting tactics like including a keyword in a website’s domain name (e.g. cheapflights.com) or URL tend to be regarded as SEO techniques from a spammier era that no longer carry weight with Google, and they ranked correspondingly low on the overall list, at #26 and #21 respectively. However, SEOs were far from unanimous on both of these fronts: the importance of keywords in domain names was the single most polarising ranking factor, with the highest standard deviation from the mean (2.7), and the use of keywords in the URL was the fourth-most divisive ranking factor, with a standard deviation from the mean of 2.59.

But where SEOs might believe that keywords in a domain or URL are unlikely to carry weight with Google, they still believe that keywords have their place in the content, title, and metadata of the page. “Exact (or near exact) use of the searched-for keywords in the content, title and meta data of the page” was ranked sixth overall in importance as a ranking factor, with relatively high levels of consensus among SEOs (it was the 7th-most-agreed-upon ranking factor).

What about links, the fuel for Google’s first and most well-known algorithm, PageRank? SEOs ranked both the quality of sites and pages linking to a particular page and the quantity and diversity of linking websites highly in terms of overall importance, at #2 and #7 respectively. They were also fairly well-aligned on these points, with both ranking factors scoring in the top ten in terms of relative agreement among SEOs.

However, SEOs were less convinced of the importance of link anchor text – both in links pointing to the page and in links to other pages on the domain – or of external links used on a page to a high ranking on Google. “Anchor text of links pointing to the page” was ranked 15th in terms of overall importance by SEOs, while “Anchor text of links to other pages on the domain” (i.e. internal links) ranked 20th, and the use of external links ranked second from bottom.

As an aside, I find it surprising that internal links didn’t feature more heavily in the list of ranking factors: while they’re mentioned in the context of anchor text, there was no mention of the quantity of internal links as a potential ranking factor, or anything regarding overall site structure.

In his personal analysis of the survey findings, Rand commented that “the perceived value of anchor text is diminishing”, noting that the anchor text of links had been a mainstay in the top few ranking factors in previous iterations of the survey, but it is now no longer in the top 10. Rand himself believes that this assessment by SEOs is “probably incorrect as an absolute assessment”, but that Google has indeed been moving away from an over-reliance on link anchor text over the past two decades.

Illustration of a link

Interestingly, despite their belief in the diminishing importance of link anchor text, SEOs also don’t give much credence to linkless mentions, or unlinked mentions, as a ranking factor. Linkless or unlinked mentions are exactly as they sound: mentions of a brand name that don’t contain a direct link to the brand website, but which Google is nevertheless able to parse as a ranking signal, as confirmed by Google’s Gary Illyes in a closing keynote at Brighton SEO in September 2017. (Bing has also confirmed that it has the same capability).

In spite of this confirmation, SEOs ranked linkless mentions 22nd out of 26th in terms of importance to Google’s overall algorithm. However, there was also a fairly high variance of opinion among SEOs as to the importance of linkless mentions, with a standard deviation of 2.28, making it the 8th most divisive ranking factor. Those SEOs who gave linkless mentions a low importance rating may either not believe that Google truly takes them into account, or not believe that they contribute much to the overall ranking of an individual webpage.

There is also little doubt that SEOs believe both links and keywords now come second to the accuracy and relevance of content in 2019. Relevance of overall page content came top as a ranking factor by quite a wide margin, with an average importance of 8.52 compared with 7.87 for the next contender, quality of linking sites and pages. The other 25 ranking factors were all much more tightly clustered together in terms of their average rating. “Content’s accuracy with accepted facts” was also ranked highly by SEOs at #8 on the overall list, and “Use of query-relevant entities in page content” only just missed making it into the top 10 ranking factors, by a margin of 0.09.

The only content-related factor that SEOs don’t seem persuaded by is quantity of content: “Total amount of on-page content” ranked 17th out of 26 overall with an importance rating of 5.69. Content freshness is also not believed to be that persuasive in 2019, and was ranked 14th out of 26, with an average importance rating of 6.37. Content freshness was likely to be much more persuasive as a ranking factor seven to 10 years ago, when the act of publishing content itself was more noteworthy, and Google needed a means of distinguishing regularly-maintained and up-to-date websites from defunct and outdated ones. Now, relevance and accuracy are more persuasive than recency, and evergreen posts from four or five years ago can still rank highly for many search terms.

How much has SEO changed in 10 years?

SEO in 2019 is a very different industry to 2009, when tactics that we would consider to be dubious today were far more commonplace, and things like Google algorithm changes and manual penalties could have wild effects on a business’s revenue, sometimes causing it to sink altogether.

But were the priorities of SEOs in 2009 completely different from 2019? A look back at Moz’s Ranking Factors Survey in 2009 suggests not. While some of the ranking factors that SEOs considered all-important in 2009 have fallen out of favour, many are still considered important in 2019.

In 2009, SEOs considered “Keyword focused anchor text from external links” to be the most important ranking factor for SEO – as we’ve established, that has changed considerably. However, second and third on the list were “External link popularity” and “Diversity of link sources”, which map to the combined factor “Quantity & diversity of linking websites” in the 2019 survey. SEOs in 2019 still ranked this highly, at #7 out of 26.

Fourth highest in 2009 was “Keyword use in the title tag”, which doesn’t map exactly to any of the ranking factors SEOs were surveyed on in 2019 – although it forms part of the expanded ranking factor “Exact (or near exact) use of the searched-for keywords in the content, title, and meta data of the page”, which ranked sixth in 2019. Finally, the fifth most important ranking factor in 2009 was “Trustworthiness of the domain based on link distance from trusted domains”. This maps to “Link authority of host domain” in 2019 as well as to “Page’s perceived expertise, authority, and trust”, which ranked 9th and 10th on the 2019 list respectively.

It goes to show that even over the course of ten years, some of the fundamentals of SEO remain exactly the same, even as other ranking factors such as mobile friendliness demand SEOs’ attention. We can see how the role of content in SEO has shifted and changed as the internet has changed, while certain tactics like keyword stuffing and link spamming have receded as Google penalises them – but the underlying principles, that the content should match the keywords of a search and that a high number of diverse, trustworthy links pointing to a page will help it rank, remain unchanged in the eyes of SEOs.