How Google Might Personalize Search Results Outside of Personalized Search

Not long ago, during a search at Google, a message at the top of the search results told me that my results were,

“Customized based on recent search activity.”

A link next to that message provided more information, telling me that if I signed into my Google Account, I might see “even more relevant, useful results,” based upon my “web history.”

During another recent search, a similar message appeared telling me that my results were based upon my location, with the results biased towards Philadelphia, which isn’t too far away.

I’ve been wondering since what it is that Google is considering when it makes changes to my search results like that. The major commercial search engines act as an index to the Web to many people who rely upon them when looking for information online.

Imagine an index that changes for every searcher.

What might that mean to searchers and to the site owners who hope that search engines will help people find their pages? What information might Google be looking at when it customizes search results based upon “recent search history,” or the location of a searchers?

We may have started receiving some clues…

Last week, Google was granted patents describing how they could change the order of pages in search results based upon the preferred language of a searcher, or by what appeared to be their preferred country for results. Both patents were filed roughly about the same time, back in 2003, and both show ways of personalizing search results for searchers based upon some information that could make the results they see more meaningful to them.

So Google might change the ordering of your search results based upon which country and which language you might “prefer” to see search results in, and it might even provide different search results to different searchers based upon an even wider range of differences between those searchers.

Another patent granted to Google this week, also originally filed in 2003, looks at different “populations” that a searcher may be part of so that it can reorder pages within search results based upon user behavior of those populations. User behavior information such as what pages members of those populations click through when presented with search results. The patent is:

Methods and systems for improving a search ranking using population information
Invented by Simon Tong and Mark Pearson
Assigned to Google
US Patent 7,454,417
Granted November 18, 2008
Filed September 12, 2003

Abstract

Systems and methods that improve search rankings for a search query by using data associated with queries related to the search query are described.

In one aspect, a search query is received, a population associated with the search query is determined, an article (such as a webpage) associated with the search query is determined, and a ranking score for the article based at least in part on data associated with the population is determined.

Algorithms and types of data associated with a population useful in carrying out such systems and methods are described.

What is Population Information?

According to the patent, after a searcher performs a search, and a search engine retrieves a listing of pages in response to that search, ranked in order of relevance, the search engine might then look to see if there is a population signal associated with the searcher, and reorder the search results that searcher is shown.

We are told that population information about searchers might broadly include such things as:

  • The locations of users,
  • The populations with which users are associated, and;
  • Information about groups with which users are associated.

Location information might include:

  • a continent,
  • a region,
  • a country,
  • a state,
  • a county, or;
  • a city.

Populations with which users may be associated might be based upon:

  • a gender,
  • a demographic,
  • an ethnicity,
  • a continent,
  • a region,
  • a country,
  • a state,
  • a county, or;
  • a city.

An example of a population with which searchers could be associated might be age ranges of those searchers, such as “under 18 years old,” “18-24 years old,” “25-34 years old,” “35-49 years old,” “50-62 years old,” and “over 62 years old.”

A slightly different way of diferentiating searchers might involve looking at “groups” with which searchers are associated, based upon things such as:

  • a gender,
  • a demographic group,
  • an ethnic group,
  • persons with a shared characteristic,
  • persons with a shared interest, and;
  • persons grouped by a predetermined selection.

.

An example provided by the patent of groups with which searchers can be associated with can be identified as “all persons interested in collecting ancient shark teeth,” and “all persons not interested in collecting ancient shark teeth.”

Self Identification and Automatic Identification Data

A search engine can gather some information to use in collecting signals about what populations we might be members of by looking at browser settings and other information independent of what we search for, by looking at what the patent refers to as “self identification-type data” and “automatic identification-type data.”

Self identification-type data can include such information as user registration data when you register for something like personalized search or other applications, user preference data such as a preferred language that you might like to see search results in, other user selected data.

Automatic identification-type data may include information collected in other ways, such as:

  • the Internet protocol address of a searcher’s location,
  • default data obtained from a searcher’s browser application program,
  • cookies, and;
  • other data collected from a searcher’s application program when the searcher’s application program interacts with a search engine.

Example

When someone in Japan types in a first search query such as “boating,” the search engine knows that the search comes from Internet protocol address located in Japan, and may see that the searcher has their browser set to a Japanese language preference. When that searcher starts selecting search results, they may start choosing pages that are in the Japanese language.

The search results that they are shown may be reordered to show them search results based upon that automatically collected information as well as the selections they make when searching. The searcher in Japan looking for [boating], may be shown search results such as “boating.co.jp,” which may be a good match for the population data that Google has collected about the searcher.

User Behavior and Population Information

If a searcher is indentified as belonging to a specific population, that searcher’s search results may be reordered based upon information about how other members of those populations interacted with web pages while searching or browsing the Web.

Our searcher in Japan, looking for pages about boating in Japanese may have his or her search results reordered based upon which pages other searchers from Japan looking for Japanese language results selected within search results.

The reordering of those search results could be based upon other activities of those searchers who share similar population signals.

Populations and Sub-Populations

Sometimes there isn’t very much user behavior activity collected for people in certain populations. That’s where the idea of sub-populations can be used to broaden the reordering of search results based upon population information.

People who live in Paris are a sub-population of people who live in France, and people who live in France are a sub-population of people who live in Europe, who are a sub-population of people who live in the World.

If there isn’t much user behavior information collected for a specific query term for people who are members of the population of Paris, the search engine might then look at the amount of user behavior information for people who are members of the population of France, and then of Europe. If there isn’t much, or no user behavior information at all for a sub-population, information from a higher population might be used to reorder search results.

Any user information about members of the population of Paris might be given a greater weight then user behavior information from the larger population of France, and even more than the weight given to the even larger population of Europe.

Clickthroughs and Other Data

A clickthrough is the selection by a searcher of a page listed in search results in response to a particular query. These searcher clicks aren’t the only kind of data that a search engine may collect that might indicate that people are interested in certain pages when searching for specific queries.

Some other information that a search engine might consider when associating pages with members of different populations might include:

  • How often a particular URL, document, or web page is shown in response to a search query;
  • How many times a particular search query is asked by users from a particular location;
  • How many times a particular search query is asked by users from a particular population;
  • How many times a particular document is selected by users from a particular location,
  • How many times a particular document is selected by users from a particular population;
  • How many times a particular document is by selected by users for a particular search query;
  • The age or time a particular document has been posted on the Web, and;
  • The identity of a source of a particular page on the Web.

Conclusion

The idea of grouping searchers into different populations, and showing those searchers lists of results for a query in an order based upon past user behavior of other members of that population means that the search results you see may be very different than the search results that I see, especially if we are from very different populations – different countries, preferring different languages, showing different interests in what we tend to search for, and so on.

The patent refers to a “smoothing factor” which “reflects how much data is needed to trust a click signal.” This means that if there isn’t very much user behavior information collected for a specific population regarding a specific query, that information might not be used to reorder the search results that you see.

What this might mean for searchers is that Google might show you results in an order based upon what it thinks might interest you the most, based upon whatever information it can collect about you, and based upon what people whom it believes share population information with you found interesting enough to look at in the past.

What this might mean for site owners is that the search engine might rank your pages higher for some populations of searchers and lower for other populations of searchers based upon a wider range of information than just whether your site is relevant for specific query terms.

Share

26 thoughts on “How Google Might Personalize Search Results Outside of Personalized Search”

  1. An enjoyable read, I think you are spot on.

    All this could be percieved as a worry for us, the great Internet may not actually be the utopia of free choice but indeed a more narrowing experience over time.

    I wrote about Ideological Amplification on a similar thought.

  2. Hi Chris,

    I enjoyed reading your article on Ideological Amplification, and it covers one of my greatest fears when it comes to search engines making “recommendations” or “decisions” on what to show a searcher based upon whether or not they fit a certain profile, or are members of certain populations.

    I want to make decisions for myself rather than having them made for me based upon what other searchers have selected, and I want to know what the assumptions and thoughts are behind recommendation systems that are making decisions on my behalf. That’s part of the reason why I spend so much time digging through search related patents.

    It’s a little distressing having an algorithm or a machine learning system determine what is relevant, and what isn’t, and not be able to see the mechanism behind the math. I would rather look at more pages, than be told that what I was looking for was contained in something recommended by an algorithm that isn’t transparent.

    A paper that has been around a while, that you might find interesting if you haven’t seen it already is Shaping the Web: Why the politics of search engines matters (pdf). What you wrote above reminded me of this passage:

    If trends in the design and function of search engines leads to a narrowing of options on the Web—an actual narrowing or a narrowing in what can be located, the Web as a public good of the particular kind that many envisioned is undermined.

    The ideal Web serves all people, not just some, not merely those in the mainstream. It is precisely the inclusivity and breadth that energized many to think that this technology would mean not just “business as usual” in the electronic realm, not merely a new tool for entrenched views and powers.

    The ideal Web would extend the possibilities for association, would facilitate access to obscure sources of information, would give voice to many of the typically unheard, and would preserve intensive and broadly inclusive interactivity.

  3. Personalized search is really taking away from the importance of tracking keyword/phrase rankings. The more personalized search becomes the more important focus on analytical data will matter.

  4. well , this will make the whole SEO game so difficult. However the new is becoming playfield for blackhatters day by day.

  5. Hi Bill,

    All this stuff on population e.g. location, state, county will all be derived from your Google account as opposed to browser data etc. Even more so, if you signup for Google Checkout, Google has a lot of more accurate data it can then use.

    Just a thought.

  6. Hi Chris Hornak ,

    You’re right – paying attention to analytics is more important than ever, if you’re concerned about how people find your web site. That doesn’t mean that keyword research itself is less important, but it does mean that tracking rankings is much less so. I believe that has been true for a few years now.

    Hi Rycroft,

    This doesn’t make SEO more difficult, but rather different. Search engines, like the rest of the Web, are evolving. I think this evolution may make it more difficult for people to use tactics to manipulate search engines, rather than less.

    Hi Nobody,

    When I was considering a title for this post, I thought of how the personalization described in this patent might differ from the personalization that Google might have to offer when someone was logged into personalized search, and Google had access to information that person shared through signups to Google services, collections of web histories through toolbars, and other services.

    The major commercial search engines all have ways to gather significant amounts of information about people who use their sites, including the information that they collect when someone creates a Google Account and uses Gmail, Alerts, Bookmarks, Google Checkout, and many of the other services that Google (or Yahoo, or Microsoft) offers.

    But the personalization described in this patent looks at information can be gathered without a Web History, or a profile page, and can avoid some of the problems that might be associated with relying upon information that is only available when someone is logged into Google.

    For example, consider that many people share computers or IP addresses in a home or office setting. Google now has a “Sync” function in their toolbar, so that if people sign into their Google Account at a computer that is shared, it will show them their bookmarks and other applications that are uniquely tailored to them. But people often don’t sign into something like that to perform a search. They may be searching at a computer that they haven’t used before at a school or a library or an intenet cafe or coffee shop.

    This patent also describes using locations, and with the increase in use of mobile phones that can access the web, people may not be searching from the locations listed in their Google Accounts, especially when they are traveling. A relevant search result may be from the location they are at, rather than from the location associated with their Google Account. The information isn’t any less accurate, but it might not be relevant if it relies upon Google Account information.

    There are a few other interesting aspects of this patent to consider.

    One of them is that the patent was originally filed in 2003, which means that it comes before some of the work that Google has done on personalized search and collecting Web histories. It considers ways to present personalized results without the existence of information that rely upon web histories and information supplied by searchers. As I noted at the start of this post, the notifications that are shown to people about their search results being modified by the search engine let them know that their results are modified based upon “recent search activity,” and they can get even more personalized results if they log in to their web histories. So, this approach allows for some personalization without being being logged in.

    Another thing that is interesting about it is that provides some insight into how web sites might be associated with different populations of searchers. That association is likely true regardless of whether a person is logged into a Google Account or not. We are shown a way that clicks (and other user behavior data) from different populations can influence which sites may be promoted or demoted in the ordering of search results.

  7. Excellent article on Personalization of Search.
    Thank you.

    It can be challenging to get people to understand how things have changed and they will continue to change.

    Traditional SEO people are already arguing with their clients about conflicting rankings and some folks are wanting to build a better “rank checking tool” thinking that the old tools are broken.

    They don’t realize the problem is not with any of the tools.

    We often hear people talk about keeping a closer eye on their analytics as a way to solve the challenge. My question is that shouldn’t you be doing this anyways?

    Checking analytics will tell you if you have a problem or whether you are making any sales or not, but they won’t tell you what you need to do to fix the problem.

    Also comments about keyword research no longer being important, (some folks seem to be lost on this) is not true. In fact keyword research will be more important than ever. Especially in the case of supporting terms.

    All in all, personalization is going have an enormous impact in the quality of user search. I also think it will ultimately clean up a lot of the SEO firms that have sort of operating “fly by night” so to speak.

    Distinguished engineer Mike Moran has stated that in his opinion, that “Personalization is one of the biggest changes since the days of PPC was introduced.”

    Another excellent resource on the topic of Personalization (the best we’ve come across) is from Search Engine Academy Trainer Mike Marshall. Michael is a Mathematician and expert in linguistics and Latent Semantic Indexing among other things.
    Michael actually does SEO training for the US Patent office.

    Keep up the good work Bill.
    Great stuff!

    John Alexander
    Directror, Search Engine Academy

  8. Hi John,

    Excellent points. Thank you.

    Rank checking tools have always been of limited utility – while you can target pages for specific keyword phrases, there are also a wide range of other terms that a page can rank for if it is well written, informative, and on topic. Analytics tools can provide a greater insight into the actual traffic that arrives at a page, and if used right can provide some ideas for developing future growth.

    The impact of personalization in search is that people who own web sites need to understand their audiences, and communicate in words that audience expects to see on the pages of the site, and will search for at search engines.

    Keyword research is more important than ever before, and with it is a requirement that the person conducting that research understand the many different audiences that a site might be created for, so that they can create content that is truly relevant to the needs of those audiences.

  9. On one hand I see how personalized search is a natural evolution in our digital world of increased customization, however I share your concern for what seems to be a “narrowing of options”. Ultimately it appears to be contrary to the long tail of choice that we are coming to expect more and more. Maybe not any time soon, but Google won’t be able to keep the discovery process within their search results pages forever. There’s just too much information out there.

    Great post!

  10. There are a lot of assumptions made by people in general about the way “rankings” have traditionally worked.

    While Google’s personalization filter is already causing some misunderstandings between SEO professional firms and their clients, it is not really a bad thing as a whole for “user experience” and while Google is leading the way, the other majors will be following too.

    Traditional thinking was or use to be:
    “What position do I “rank” for a specific keyword phrase on a linear scale.”

    Many people do not realize that search engines do not grade on a linear scale anyways. Neither do search engines grade keywords on an absolute scale.

    But with respect to “personalization of search” we are not dealing with a new algorithm either.

    What you need to keep in mind is that there is a shuffling of results that occurs by a filtering process for the top 100 results or so, based on the worlds personal Web history.

    New Thinking Applies:
    What percentage of your pages are getting top exposure or visibility (not based on traditional concepts of linear ranking) for the most effective keyword phrases – for what percentage of the time and to the “most ideal” searching audience.

    To me it is never about trying to focus too much on what Google is doing, but people need to understand the challenge before tackling the solution.

    We try to help people understand the challenges, the reasons why something works the way that it does. Once you have understanding of the theory, there are always more than one solution to any challenge. This is no different.

    There are certain things you will never discover from Google.

    They have to protect their information and they have a number of steps in place for the purpose of protecting it.

    The good news is you do not have to be obsessed with Google or any search engine’s algorithm. Now with personalization, new rules are going to apply and once you are understanding them and applying them, the nightmares all go away.

    The challenge is that there are so many people that learn the “how to” or the what to do” factor, but they never learn the reasons why. Therefore, people who don’t stay up to date with their skills, are going to find it extremely difficult to understand what has and is transpiring with expect to the changes.

    Here is another question for you?
    Do you think that the personalization of search is also going to affect the PPC rankings? I know a few very respected researchers that think so.

  11. Hi John,

    We know that Google will provide paid search results that are based upon previous searches, without a searcher being logged into Google’s Personalized search. For example, I searched for the term [men's shoes] and saw these results:

    Google paid search results in a search for men's shoes

    I then performed another search for the term [golfing] and received paid search results for men’s golfing shoes:

    Google paid search results in a search for golfing

    Google has published more than a couple of patent filings related to personalized paid search results, too.

  12. Hey Bill,

    I love it and I am in agreement.

    I posed the question for the sake of the readers who may not be quite as well informed.

    You have a great topic and great conversation thread going here. Light years ahead of some. Great stuff.

    Best,
    John

  13. Hi Will,

    Thanks. You raise some very interesting points:

    Ultimately it appears to be contrary to the long tail of choice that we are coming to expect more and more. Maybe not any time soon, but Google won’t be able to keep the discovery process within their search results pages forever.

    That could be one of the risks that reordering results based upon personalization might present. Often people search for things that they don’t know much about, and personalization may have the potential to keep them from discovering new information by presenting results that more match what they’ve looked at before than what they may be looking for at the time of a search.

    Diversification and personalization may aim at different purposes – too much personalization may prevent diverse results from appearing as highly. I have seen a number of patent filings that do focus upon providing a diverse set of search results, based upon looking at things like categories that might be associated with each of the pages in the search results. Balancing different goals in search results might be difficult.

  14. Hi John,

    Thanks for your kind words. I really appreciate all of the input that people provide here.

    I’m glad you raised a question about personalization and paid search. Thanks. What I posted was one example, and I’m sure that there are other ways in which personalization may influence rankings of paid search results. It’s something that people using paid search should be thinking about.

  15. Pingback: The Search Visibility Plan, Not Your Average SEO Campaign
  16. I wonder how the population and country profile will effect websites that offer international products i.e. international real estate. Does this mean a search undertaken in Canada for property in Spain will give a result where Canadian websites with Spanish property are given to the user and not Spanish based websites?

  17. Hi Nick,

    Really good question.

    One of the recent patents granted to Google discussed country preferences that searchers might have, and how the search engine would try to boost search results for pages that might be from those countries, and possibly slightly reduce rankings for sites that are from countries that aren’t preferrred.

    In that patent, it mentioned that some sites aren’t tied to specific countries, but rather are world wide in orientation, and probably wouldn’t be affected. The types of sites that you mention, such as a Canadian site offering Spanish property sounds like it might be considered world wide in a profile for it.

  18. I hope not, this personalised adsense was quite the blunder. If someone was searching alot for cars, then they found what they wanted, then they were interested in dogs and visit dog websites. They find on that site adverts for cars wen they are now interested in dogs, useful? no, webmaster loses loads of revenue, Google loses webmaster as he moves onto a better ad system.

  19. Hi Tom,

    You’re right. Ideally, if there is very little in common between topics, than a personalized advertising system shouldn’t attempt to combine the two topics.

    I still think back to the late 90s when Altavista was one of the top search engines, and on a search for “Dalai Lama” one of the search ads offered me the Dalai Lama at the lowest price on the Web. I started using Google a lot more after that…

  20. Research shows the average internet users behavior is very random and unpredictable, so to base a system on it is ludicrous! Google should have allowed webmasters to at least choose what kind of ads to display in terms of personalised or what they were before.

  21. Hi Tony,

    I’ve been looking at a lot of research papers from University researchers and from the search engines which have been identifying some patterns of behavior for searchers over different search sessions. The different metrics that are described in the patent aim at trying to idenrify user behaviors that might follow a pattern. It’s possible that some behavior might carry some level of predictability. It’s an area that Google does seem to be testing and exploring.

    This particular patent filing focuses upon the ads that show up as sponsored results next to web search results on Google’s pages rather than on sites that might show Google ads as part of Google Adsense. Google’s approach to Adsense does seem to focus upon serving ads that are relevant to the content displayed upon those pages, and allows site owners to block advertisers.

  22. Having worked for years in a big box SEO house, the personalized search forced a change in the game. I recall one trade show in NY where some of the SEO pundits shared a common mantra “ranking is dead”. We knew that SERPS were being customized based on a Google account if you were logged in and we later found out that even when not logged in, our searches were still getting personalized somehow. We later assumed it was based on IP.

    The “SEO experts” panel ask the audience how many people who worked as professional SEO’s still gave out ranking reports to clients and nearly everyone still raised their hands. Was ranking “dead”? Clearly not. But our methods were forever changed and on top of that we had to educate many of our customers who perceived their value from our service based on their ranking before anything else.

    I think personalized search benefits less technically inclined searchers but I want access to the pure unbiased data. I just noticed Yoast created a plugin that will de-personalize searches from a FF or IE browser and I’m going to try it today. If I had a dime for every time I added &pws=0 to the end of a search string to eliminate the bias I could take my kids to Disneyland.

  23. Hi Mal,

    Some very interesting points.

    Ideally, traffic increases should come from both head terms and long tale terms, and ranking reports aren’t going to necessarily capture all of those increases in an effective manner, especially with personalization impacting the rankings that potential visitors view.

    Being able to see changes reflected in analytics programs can be more meaningful, and also lead to ideas for additional changes that can have a positive impact. I’m a big fan of educating clients, and helping them to understand why recommendations are suggested, and how those can be reflected in analytics reports. – I’d much rather spending time on that then on rankings reports, and going through analytics can often provide suggestions on additional changes to a site.

    I haven’t tried the plugin from Yoast, but I agree with you that it’s worth looking at those results.

Comments are closed.