Moz công bố bảng điều trang tiêu chí xếp hạng

Search Engine Ranking Factors Survey

Moz’s 2013 Search Engine Ranking Factors presents survey data from over 100 search engine professionals and provides insight into the inner workings of the future of search. The following data represents the opinions of our respondents of the various weighting of factors thought to be used (or not used) in Google’s search algorithm.

1. The Overall Algorithm

According to our survey respondents, here is how Google’s overall algorithm breaks down.

ValuesDomain-Level, Link Authority FeaturesPage-Level Link MetricsPage-Level Keyword & ContentBasedPage-Level, Keyword-AgnosticFeaturesDomain Level Brand MetricsUser Usage & Traffic/Query DataPage-Level Social MetricsDomain Level Keyword UsageDomain-Level, Keyword-AgnosticFeatures02.557.51012.51517.52022.5

Notes on the survey’s responses regarding the overall algorithm:

  1. Links are still believed to be the most important part of the algorithm (approximately 40%).
  2. Keyword usage on the page is still fundamental, and—other than links—is thought to be the most important type of factor.
  3. SEOs do not think social factors are important in the 2013 algorithm (only 7%), in contrast to the high correlations.

 

2. Domain Level Brand Metrics

These features describe elements that indicate qualities of branding and brand metrics.

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesSearch volume for the brand/domainQuantity of co-occurrence Keyword +Brand across webQuantity of mentions of thebrand/domain on social sitesQuantity of +1s to Google+ Brandpage/Places pageVolume of toolbar/browser usage datafor the domainQuantity of unique feeds that contain amention of the brand name of the sitePopularity of business’s official socialmedia profiles (Twitter, Facebook,Foursquare)Citations for the domain in WikipediaWebsite is listed frequently on Linkedinprofiles as an employer012345678

Comments on Domain Level Brand Metrics

Tim Grice
“We are seeing a strong shift towards co-occurrence, and brand + keyword search volumes. Instead of anchor text, your link strategy should focus on developing opportunities for people to talk about your brand alongside your main product or service.”
Rob Kerry
“Social signals are still in their infancy. Unless your brand exposure is enough to reach a significant market share, concentrate on traditional PR and branding for old-school signals.”

 

3. Domain Level Keyword-Agnostic Features

These features relate to the entire root domain, but don’t directly describe link or keyword-based elements. Instead, they relate to things like the length of the domain name in characters.

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesUniqueness of content across thewhole siteFreshness of content on the siteDomain has never been penalizedAggregated CTR from Google SERPsfor the domainAge of DomainQuantity of error pages crawled on thesiteUse of Responsive Design and/ormobile optimizedAggregated page load speed for pagesof domainDomain is associated with highauthority author(s) through Google+Aggregated “Dwell Time” or “LongClick” metrics for domainThe number of 404 pages linked tofrom the domainQuality of other sites hosted on thesame c-block of IP addressesUse of rel=”publisher” on domainPresence of hyphens used in domainnameNumber of characters in domain nameLength of time until domain nameexpires0123456789

Comments on Domain Level Keyword-Agnostic Features

Russ Jones
“When Matt Cutts says you will be ‘surprised’ about the level of influence of handling mobile right, it is time to start handling mobile right.”
Carlos del Rio
“The freshness of content and load speed are the most consistent factors in this group. Currently authorship metrics are being overvalued, and I expect to see this factor weaken over the near future.”

 

4. Domain Level Keyword Usage

These features cover how keywords are used in the root or subdomain name, and how much impact this might have on search engine rankings.

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesKeyword is the exact match rootdomain name (e.g. “keyword.com”)Keyword is present in root domainname (e.g. “ABCkeyword.com”)Keyword is closely related to domainname through entity associationKeyword is the first word in the rootdomain name (e.g. “keywordABC.com”)Keyword is the subdomain name (e.g.”keyword.ABC.com”)Keyword is the secondary subdomainname (e.g. “keyword.123.ABC.com”)Keyword is used in a hybrid of thedomain name (e.g. “key.word.com” or”key.wo.rd”)Keyword is the domain extension (e.g.”ABC.keyword”)01234567

Comments on Domain Level Keyword Usage

AJ Kohn
“Like it or not, having the keyword in the domain helps—sometimes more than it should.”
Dev Basu
“Exact-match domains still perform really well, despite Google updates to diminish the ‘free pass’ EMDs typically get. If I were to start a new site today I’d prefer to build a brand.com oriented domain rather than making the EMD my main site.”
Carlos del Rio
“Exact match domains are still overvalued in Google’s algorithm. I expect that we are going to see continued reduction in exact match value over the next year.”

 

These features describe link metrics about the domain hosting the page.

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

Comments on Domain Level Link Authority Features

Julian Grainger
“Context again. What are we trying to rank? How long is it useful content? Velocity can be a blessing or a curse. If your brand is global, your links should be too.”
Will Critchlow
“I suspect sentiment is coming soon, but I haven’t seen any evidence of it in use yet.”
Matt Gratt
“Google seems to be depreciating more links than ever before, so this is actually increasing the impact of getting the links that count, IMHO.”

 

6. Page Level Keyword-Agnostic Features

These elements describe non-keyword-usage, non-link-metrics features of individual pages (such as length of the page, load speed, etc.).

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesUniqueness of the content on the pageCTR from Google to the page for thekeywordFreshness of the content on the page(i.e. the time since the page was lastupdated)Page Load SpeedAuthorship is associated with a highauthority Google+ profileThe age of the page (i.e. the time sincethe page was first discovered)Pure bounce rate of the page asmeasured by return visits to the searchresults pageLength of content on the pageDwell Time or “Long Click” metricsPage contains schema.org or otherstructured dataUse of rich media (video, slides, etc) onthe pageUse of images on the pagePage contains rel=”author” markupReading level of the content on thepageUse of advertising on the pagePage contains Open Graph Data and/orTwitter Cards0123456789

Comments on Page Level Keyword-Agnostic Features

Greg Niland
“Many of these factors are intertwined. Does adding authorship markup boost your rankings? Or is it because the authorship markup added your smiling face to SERPs, which in turn drove a higher CTR? Another example is the length of content. Are you ranking because your content is long or because your long content contains extra synonyms that can’t be squeezed into shorter content? Or maybe longer content keeps users on your site longer, which can reduce bounce rates? It is hard to isolate a single ranking factor, but to be honest, you don’t need to be that precise. It’s like baking a cake: It doesn’t matter how much of one ingredient is in the recipe. It matters much more that the sum of the ingredients taste good.”
Scott Smith
“Structured data’s importance varies greatly by vertical. If I could do one thing to help every webpage out there rank better these days, it would be to add a video.”
Ruud Hein
“Approach your content from a conversion optimization point of view; that’s how Google is doing it.”

 

7. Page Level Keyword Usage

These features describe use of the keyword term/phrase in particular parts of the HTML code on the page (title element, H1s, alt attributes, etc.)

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesKeyword present in the titleelement/tagKeyword appears in the main “content”area of the pageDegree of optimization for a topicmodeling algorithm (e.g.LSA/LSI/pLSI/LDA/etc) on the entirecontent of a page as it relates to theuser’s queryKeyword is the very first word/phraseof the title elementKeyword present in the first 100 wordsof the documentKeyword present in the page’s URLstringKeyword present in the page’s HeaderTags (H1, H2, H3)Keyword present in the anchor text ofan on-page linkKeyword present in the alt attribute ofan image on the pageKeyword “density” (percent of words onthe page that are the target keyword) isan “ideal” quantityKeyword present in bold/italic/em tag0123456789

Comments on Page Level Keyword Usage

Christine Churchill
“Good on-page optimization is STILL important. It helps Google understand what the page is about. Google won’t rank a page until it knows what the page is relevant for.”
Russ Jones
“As Google becomes better at what it does, the reliance on tangible, on-page features seems to be disappearing as fast-if not faster-than link metrics.”
Abhilash Patel
“Keyword usage seems to have decreased a bit in importance, especially relative to brand importance. In many cases we have observed industry authority webpages with little keyword usage actually outweigh more informative, keyphrase-relevant pages, presumably on brand authority.”

 

8. Page Level Social Metrics

These features relate to third-party metrics from social media sources (Facebook, Twitter, Google+, etc.) for the ranking page.

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

ValuesAuthority of users sharing page viaGoogle+Quantity of Google+ Shares and +1sAuthority of users tweeting links to apageSize of Google+ author circleSize of Google+ publisher circle (e.g.the number of people who have brandpage in circles)Quantity of tweeted links to pageQuantity of Facebookcomments/likes/shares for the pageAuthority of users sharing link onFacebook to a pageSentiment of social links and citationsreferring to the pageComments about the page on socialsites e.g. reddit, stumbleupon, etcUpvotes for the page on social sitese.g. reddit, stumbleupon, etc0123456

Comments on Page Level Social Metrics

Hannah Smith
“Sorry—I’ve not seen any evidence that’s convinced me that page-level social metrics play a part in the algorithm. I think these signals are either too easy to fake, or in the case of sentiment the search engines, just aren’t quite there yet. I think in the future there’s the potential for them to become more heavily weighted, but I don’t think the data is reliable enough yet.”
Todd Malicoat
“Expect Google to push for higher adoption of G+, and adopt more social signals as their network of users grows and becomes more reliable for important relevance data points.”

 

These features describe link metrics to the individual ranking page (such as number of links, MozRank, etc.)

Rated on a scale of importance from 1 (No Importance) to 10 (Very Important)

Comments on Page Level Link-Based Features

David Iwanow
“Link velocity seems to be coming back. If you stop building links you fall down in rankings … competitors who can keep up velocity of links can seem to rise in rankings even if the quality of the links is low.”
AJ Kohn
“Topical relevance of linking pages and domains remains the most powerful signal, and it’s only getting stronger as topic modeling is influenced by entity detection.”
Tim Grice
“Anchor text has had its day—in fact, we have been running successful strategies for years without considering anchor text. Having a strategy around number of anchor text or a specific ‘anchor text variation formula’ is just ridiculous and will lead to problems somewhere down the line.”

 

Participants were asked to share their predictions of how the following factors might change (in terms of their impact on Google’s ranking algorithm) over the next 12 months.

ValuesDecreaseStay the sameIncreaseAnalysis of a site/page’s perceivedvalue to usersAuthorship Metrics (Author / AgentRank)Influence of Structured Data in SERPsThe overall influence of Google+Social signals – shares, likes, +1’sContent readability / usability / designUsage data (CTR, dwell time, etc)Quantity of paid results on SERPsOn-Page topic modeling (e.g. LDA)Presence + prominence of advertisingvs. contentInfluence of Anchor text in externallinksThe effectiveness of paid linksInfluence of Anchor text in internallinksExact keyword match domains020406080100120140

Comments on the Future of Search

Dev Basu
“Social signals from Google+ will become more and more relevant over the next 12 months. Eventually Google+ will take equal, if not more, prominence than Facebook due to its position at the cross-section of search and social.”
Laura Lippay
“Some of the research papers from universities and/or search engines have shown them testing things like how long the window is in focus, scrolling on a page, printing from a page, and-what I think is the most interesting one (used for testing a page’s credibility)-automating the process of identifying topical experts and giving weight to the pages they visit.”
Todd Malicoat
“Don’t believe the hype of social and G+. They will increase in importance, but certainly not replace more traditional offsite factors. Inbound links and offsite equity will continue to be highly influential to search relevance when validated by other factors. Expect these offsite equity trust and authority factors to maintain a large portion of the importance that they’ve always held as the foundation of the Google algorithm.”