Since my blog post on Eli Pariser’s Ted talk about the filter bubble became quite popular and a lot of people seem to be interested in which 57 signals Google would use to filter search results I decided to extend the list from my article and list the signals I would use if I was google. It might not be 57 signals but I guess it is enough to get an idea:

  1. Our Search History.
  2. Our location – verfied -> more information
  3. the browser we use.
  4. the browsers version
  5. The computer we use
  6. The language we use
  7. the time we need to type in a query
  8. the time we spend on the search result page
  9. the time between selecting different results for the same query
  10. our operating system
  11. our operating systems version
  12. the resolution of our computer screen
  13. average amount of search requests per day
  14. average amount of search requests per topic (to finish search)
  15. distribution of search services we use (web / images / videos / real time / news / mobile)
  16. average position of search results we click on
  17. time of the day
  18. current date
  19. topics of ads we click on
  20. frequency we click advertising
  21. topics of adsense advertising we click while surfing other websites
  22. frequency we click on adsense advertising on other websites
  23. frequency of searches of domains on Google
  24. use of or google toolbar
  25. our age
  26. our sex
  27. use of “i feel lucky button”
  28. do we use the enter key or mouse to send a search request
  29. do we use keyboard shortcuts to navigate through search results
  30. do we use advanced search commands  (how often)
  31. do we use igoogle (which widgets / topics)
  32. where on the screen do we click besides the search results (how often)
  33. where do we move the mouse and mark text in the search results
  34. amount of typos while searching
  35. how often do we use related search queries
  36. how often do we use autosuggestion
  37. how often do we use spell correction
  38. distribution of short / general  queries vs. specific / long tail queries
  39. which other google services do we use (gmail / youtube/ maps / picasa /….)
  40. how often do we search for ourself

Uff I have to say after 57 minutes of brainstorming I am running out of ideas for the moment. But this might be because it is already one hour after midnight!

If you have some other ideas for signals or think some of my guesses are totally unreasonable, why don’t you tell me in the comments?

Disclaimer: this list of signals is a pure guess based on my knowledge and education on data mining. Not one signal I name might correspond to the 57 signals google is using. In future I might discuss why each of these signals could be interesting. But remember: as long as you have a high diversity in the distribution you are fine with any list of signals.

If you like this post, you might like these related posts:

  1. Google uses your Location to personalize search results So today I’ll start my research about the 57 signals...
  2. Algorithmic Information Filter from Eli Pariser’s TED Talks Just today an interesting story came up on a German...
  3. Google Video on Search Quality Meeting: Spelling for Long Queries by Lars Hellsten Amazing! Today I had a discussion with a coworker about...
  4. My Blog guesses your name – Binary Search Exercise for Algorithms and data structures class Binary Search is a very basic algorithm in computer...
  5. IBM’s Watson & Google – What is the the difference? Recently there was a lot of news on the Web...


Tags: , , , , , , , ,

39 Comments on What are the 57 signals google uses to filter search results?

  1. [...] find my list of almost 57 signals google might use to filter [...]

  2. [...] Mohrrüben zu tun haben.Drittens kann ich nicht beeinflussen, was genau die Filterkriterien sind. Google wertet angeblich 57 Signale aus, die mich in einem Raster einstufen lassen. Welche das sind und wie ich sie ändern kann, [...]

  3. [...] today I’ll start my research about the 57 signals Google uses to personalize search results. To verify that Google uses your Location to tailor search results was an easy score. After the [...]

  4. [...] zal krijgen: het soort pc, welke browser die je gebruikt, de locatie van waar je surft, enzovoort. (hier wordt er een gokje naar deze 57 signalen gewaagd)  En ook The Washington Post, The Huffington Post en zelfs de grootste nieuwsservice op het [...]

  5. [...] (got that list over at  What are the 57 signals google uses to filter search results?) [...]

  6. TomB says:

    Just throwing out some ideas:

    -the ISP
    -connection speed
    -Maybe how long the computer has been powered on
    -is google the homepage
    -do you use the google toolbar
    -the referring sites (so which page you were looking at before googling)
    -is safe search turned on
    -maybe web/browser security level
    -there’s gotta be something involving porn, so maybe whether you search for porn
    -the number of different users from your IP address
    -time of day?
    -whether you use google at work or at home

    • René Pickhardt Rene says:

      hey Tom thanks for your suggestions. I am busy right now but I will extend my list with the suggestions you made that sound reasonable to me.

  7. [...] only travel links, Pariser was served mostly political links. (Pariser guesses that Google uses 57 signals to filter search results.) So he came up with the term "filter bubble" – your personal universe [...]

  8. Ken Brown says:

    - do you usually select the first hit? Or second? Or go down the page?

    - how quickly do you come back to Google?

    - how far down the pages do you go? (e.g. I often jump straight to page anywhere between 2 & 10 if the first hits all seem to be advertising sites. Or bloody Amazon, which grabs dozens of search hits including the title of a book)

    - do you use image search?

    - do you turn safe search off?

    - do you use quotes?

    - if Google suggests a new search do you take it or continue with your own?

    • René Pickhardt Rene says:

      hey Ken
      I don’t see how your points actually extend the above list. Maybe I understand my points from a different perspective than yours. Then please explain them. The only new thing I see (besides safe search which Tom mentioned) is the “use of Quotes” this definately changes search results and of course it could also be a signal because it might help to identify the advanced user but I would guess there should be other signals.

  9. [...] Holdings Expand (Bloomberg) • What are the 57 signals google uses to filter search results? (René Pickhardt) • Sports: Lionel Messi, Boy Genius (NYT) • A Twist on Climate Change, Risk, and Uncertainty [...]

  10. I’d suggest looking at various patents for a better idea of what’s actually in use.

    • René Pickhardt Rene says:

      great Idea! That would be really interesting and probably give even more insights. Are you aware of an already compiled list of patents around Google search?

  11. [...] other fifty are guessed at by Rene Pickhardt, a self described “webscience” blogger at .  All of it is interesting reading, but the gist of it all is that our browsing behavior is [...]

  12. James Bhatt says:

    -Do you open search results in new tabs, or navigate away in the same tab? And going on from that;
    -Do you open many results in tabs/windows and then peruse them or open & then read each result immediately before going to the next?

    -Do you more often choose ‘soft’ information like wikipedia or blogs, or ‘hard’ information like news sites?

    That might suggest someone doing more some sort of research, or maybe someone who favours getting many opinions on a subject.

  13. [...] new customers because they are only pounding on people that are already interested in something as indicated by their activity?  While I don’t think the traditional demographic segmentations will hold up without [...]

  14. [...] The most interesting thing is that most of the suggested signals are useful for identifying who is doing the searching, in the absence of a name and address.  As I mentioned in my last post, see Rene Pickhardt’s site for  more detailed and interesting speculation. [...]

  15. [...] from all over the world from trying to reverse engineer the massive Google engine, here is one of the best lists generated so [...]

  16. [...] Pickhardt, a Webscience PHD student, took a crack at naming at least 40 of them. These are by no means confirmed by Google, but it’s an [...]

  17. [...] What are the 57 signals google uses to filter search results? [...]

  18. [...] Seulement sur Internet, le consommateur d’information ne connait pas forcément la qualité de cette information (panée ou pas). C’est le cas avec les moteurs de recherche, qui, il y a à peine 10 ans, nous donnaient des résultats basés sur des algorithmes logiques de qualité des sites web. Aujourd’hui ces mêmes moteurs (Google et les autres) vont nous donner des résultats basés sur la pertinence de leur classement, mais également sur des “informations personnelles”. On en compte 57 chez Google. [...]

  19. [...] of Library and Information Studies blog.  It’s a great link to a TED talk outlining the 57 signals Google uses to filer each person’s searches.  Scott’s post came right before the Times [...]

  20. [...] bei Google Deutschland, der ComputerBILD. Um die Google-Suche zu optimieren, nutzt Google 57 Indikatoren, die bei jedem Besuch untersucht werden. Neben Informationen über unsere Search History, Browser, [...]

  21. Tom says:

    I got a good one we may have missed..
    the amount of times you go to tools and delete browsing history
    and the timing of deletion (ex: do you delete after certain type of sites and searches, and do you delete more often on one particular time of the day, etc..)

    Also, the general sense of the blog when words are logged in multiple times, the tone for lack of a better term (I loved this, this is great, you are the bomb..) versus (i disagree, i dont like this, etc..)

    maybe even the changes between capitalization and punctuation, come to think of it.. internet behaviorism

    the amount of cuss words.. wow forget 57 google can easily track hundreds of them.. whos to say google doesnt notice and take offense to not giving it a capital G..

  22. [...] var vi befinner oss. Men majoriteten av dem är hemliga. Smarta människor bjuder dock på några kvalificerade gissningar.  De menar att Google bland annat tittar [...]

  23. [...] I had my first blog article or thing on the internet that became kind of viral. [...]

  24. [...] realize is the extent that Google is so customizing what it presents to me based on their 57 signals, that no two of us see the same information.  Pariser does an excellent job of presenting both the [...]

  25. [...] merkt sich ja ziemlich viel. Um etwa die Google-Suche zu optimieren, nutzt die Suchmaschine 57 Indikatoren, die bei jedem Besuch untersucht werden. Neben Informationen über unsere Search History,  den [...]

  26. [...] are the 57 filter signals?René Pickhardt makes a good guess in his blog post: “What are the 57 signals google uses to filter search results?”The ones that particularly caught my attention:Our locationDistribution of search services we [...]

  27. [...] über komplexe Algorithmen, die mich in technischer Hinsicht nicht wirklich interessieren. Aber: Auf Basis welcher User-Merkmale bestimmt sich die Zusammensetzung der Suchresultate? Die Idee einer eingehenden Analyse meines [...]

  28. [...] grâce à une série de signaux dont le nombre, selon René Pickhardt, se situerait autour de 57. Bref, avec la même équation de recherche, un militant altermondiste ne recueillera donc pas les [...]

  29. [...] PusherVery interesting topic. I did some search around and found someone who listed his guess:…I think the answer is not that far off, Google can gather information from us even when we are not [...]

  30. [...] ancient cultures, that is a signal in the practical judgment we need to make, in the same sense as Google uses signals in their searches. It’s a signal that there may be something here that we should do [...]

  31. [...] qui nous y est présentée a été présélectionnée. Pour y parvenir, Google se base sur  57 signaux différents (votre historique de recherche, votre localisation, votre modèle d’ordinateur, …) et [...]

  32. [...] of my family members, close friends, or co-workers. The number 57 came from an engineer, and this list of the possible signals Google may use comes from Rene Pickahrdt: a Webscience PHD [...]

  33. [...] selling off data to marketing sites, whom in return process your data along with the other “57 signals,” instead, they will be able to now justify the payment for added security and privacy. I am [...]

  34. [...] Information on Google’s signals is not public, though we can have a guess: [...]

Leave a Reply



Subscribe to my newsletter

You don't like mail?