Google’s top search ranking factors & CTR in SERPs

by 27 replies
27
In a Q&A session Andrey Lipattsev, a Search Quality Senior Strategist at Google, said the most important two factors are links and content. Last year they claimed that Rank Brain was the third most important factor. This is how SEL got the list of the top three. The top two are hardly surprising to any SEOer.

Brian Dean recently linked to this slide deck by Google's Paul Haahr. He seems to confirm that Google is in fact using clicks from search, CTR, and clicks back to search as a ranking factor. This is of course something that people have been suspecting for ages. They also test their own search results by doing A/B live testing. Haahr said that if you do a query you're often taking part in at least one test.

Here's Haahr's talk from SMX West:

SMX West 2016 - How Google Works: A Google Ranking Engineer's Story - YouTube
Nothing new, but a couple of small confirmations from Google.

#search engine optimization #factors #google’s #ranking #search #top
  • Banned

    Technically it's (followed links + text).

    I'm not trying to nit pick just pointing those things out because any old link isn't ranking a page (ex: nofollow link, javascript link, flash link, etc...) & content doesn't necessarily have to be articles like most IMers would assume. It's plain text that's optimized for SEO which can be any text on a webpage (ex: heading tags, alt-text, image captions, anchor-text, page title, bullet list, tables, etc...) basically any text that shows up on the text version of a webpage.

    Really I shouldn't even say webpage because it could be any file type that Google can index. They all rank the exact same way (followed links + text).

    [source]
    • Adobe Flash (.swf)
    • Adobe Portable Document Format (.pdf)
    • Adobe PostScript (.ps)
    • Autodesk Design Web Format (.dwf)
    • Google Earth (.kml, .kmz)
    • GPS eXchange Format (.gpx)
    • Hancom Hanword (.hwp)
    • HTML (.htm, .html, other file extensions)
    • Microsoft Excel (.xls, .xlsx)
    • Microsoft PowerPoint (.ppt, .pptx)
    • Microsoft Word (.doc, .docx)
    • OpenOffice presentation (.odp)
    • OpenOffice spreadsheet (.ods)
    • OpenOffice text (.odt)
    • Rich Text Format (.rtf, .wri)
    • Scalable Vector Graphics (.svg)
    • TeX/LaTeX (.tex)
    • Text (.txt, .text, other file extensions), including source code in common programming languages:
    • Basic source code (.bas)
    • C/C++ source code (.c, .cc, .cpp, .cxx, .h, .hpp)
    • C# source code (.cs)
    • Java source code (.java)
    • Perl source code (.pl)
    • Python source code (.py)
    • Wireless Markup Language (.wml, .wap)
    • XML (.xml)








    Like you said, we already knew traffic clicking back to a search query was a bad idea. If anything it's a complete fail on behalf of sales conversions. No point in doing SEO If traffic is going to bail once they reach the money page.
    • [ 1 ] Thanks
  • Banned
    I'm watching the OP video now, funny Paul mentions Pagerank at 9:22.

    Hasn't anyone told Google Pagerank is dead?
    • [1] reply
    • Why would they?? He's a Google engineer who can still see pagerank internally. Its not dead to him but its dead to everyone else because they will never again know the actual pagerank for any site on the web..
      • [2] replies
  • Banned
    Great, after I'm done packing orders and dealing with auctions I'll watch that. Seems interesting.

    Edit: awesome thanks! There's a bit of internals discussions, things we didn't know publicly before, nothing that will rank your site, but indeed very interesting.

    Edit2: Slideshow is extremely revealing. Look at slide 23 carefully, I've been suggesting to people that all the ranking factors work like that for years.

    So in a competitive SERP, does https matter? Well it's going to be factor 100, so it's worth 1/100, probably doesn't matter. Same thing with social signals, people who say "well I build a site, buy social signals, then build links" always seem to swear by the social signals, anybody who does it the other way around argues that social signals do nothing...

    Note: Theory and probably bad math ahead

    Granted we'll never know how Google weights everything out but let's say you optimize for a factor that logically can't be worth much. Like https, let's say it's factor #50. So 0.02 / 4.499205338 = 0.0044452294344251

    So that factor contributes to 0.44% of your ranking, unless we operate a massive website, that's not going to do anything. In reality it's probably much less then that. Obtaining a single quality link will be a better use of our time.
  • Thanks for posting that, it's a very interesting read, far removed from the usual.
  • I think your site have all features w3validation than you don't need above these points.
  • I don't want to chuck petrol on the blaze here guys, just sprinkle some water on it

    Can we just clarify a small point...

    Is it not correct that nofollow links, although not passing PR, do 'absorb' or dilute some of the 'juice' flowing into a page, thus the lowering the outflowing PR.

    Wasn't this introduced by Big G some time ago to prevent PR sculpting? I'm almost sure there's a Matt Cutts video on this somewhere. Thoughts?
    • [1] reply
    • Yes think it was back in 2009 some changes were made but it certainly does not stop all PR from flowing to internal pages through some "followed" navigational links to make all new internal pages real NAs. shucks even some pages with no links with next to no links got PR zero and PR internal to Google was even fractional although the toolbar only showed whole values.
      • [1] reply

Next Topics on Trending Feed

  • 27

    In a Q&A session Andrey Lipattsev, a Search Quality Senior Strategist at Google, said the most important two factors are links and content. Last year they claimed that Rank Brain was the third most important factor. This is how SEL got the list of the top three. The top two are hardly surprising to any SEOer. Brian Dean recently linked to this slide deck by Google's Paul Haahr. He seems to confirm that Google is in fact using clicks from search, CTR, and clicks back to search as a ranking factor. This is of course something that people have been suspecting for ages. They also test their own search results by doing A/B live testing. Haahr said that if you do a query you're often taking part in at least one test.