Month: May 2020

Search Suggestions from Previously Submitted Searcher Queries

I came across an interesting Search Engine Land post last week. It inspired me to search and see if I could find a patent that might be related to it from Google:

Google is suggesting searches based on users’ recent activity

I tried to reproduce the search suggestions that were being shown to the author of the Search Engine Land article, but Google would not return those to me. Google may be experimenting with a limited number of searchers rather than showing those results to everyone. I did find a patent discussing search suggestions that were similar.

When Google shows a search suggestion about something you may have searched for in the past, that predicted suggestion is likely related to a patent I’ve written about before, Autocompletion using previously submitted query data.

I wrote about that patent being updated in a continuation patent, but hadn’t provided much in the way of details about how it works at: How Google Predicts Autocomplete Query Suggestions is Updated.

There are some interesting parts about how search suggestions are identified and ranked, which inspired me to write this post.

Search Suggestions Based on Previously Submitted Query Data

The description of this patent starts off by telling us that it is about: “using previously submitted query data to anticipate a user’s search request.”

That pinpoints that Google has a long memory, and it remembers a lot about what someone might search for.

This patent description also includes a lot of the assumptions that search engineers make about searchers (often an interesting reason to read through patents). Here are some from this patent that are worth thinking about:

Internet search engines aim to identify documents or other items that are relevant to a user’s needs and to present the documents or items in a manner that is most useful to the user. Such activity often involves a fair amount of mind-reading–inferring from various clues what the user wants. Certain clues may be user-specific. For example, the knowledge that a user is making a request from a mobile device, and knowledge of the location of the device, can result in much better search results for such a user.

Clues about a user’s needs may also be more general. For example, search results can have elevated importance, or inferred relevance, if a number of other search results link to them. If the linking results are themselves highly relevant, then the linked-to results may have particularly high relevance. Such an approach to determining relevance may be premised on the assumption that, if authors of web pages felt that another web site was relevant enough to be linked to, then web searchers would also find the site to be particularly relevant. In short, the web authors “vote up” the relevance of the sites.

Other various inputs may be used instead of, or in addition to, such techniques for determining and ranking search results. For example, user reactions to particular search results or search result lists may be gauged, so that results on which users often click will receive a higher ranking. The general assumption under such an approach is that searching users are often the best judges of relevance, so that if they select a particular search result, it is likely to be relevant, or at least more relevant than the presented alternatives.

A Summary of the Search Suggestions Process Based on Previous Submitted Queries

Like most patents, the Description for this one starts out with a summary section that provides an overview of how the process defined in the patent works. It is followed by a “Detailed Description” section that goes into more depth and provides details about how search at Google works, and how specific aspects of search at Google power this search suggestion process. So read about how search suggestions might be provided based upon user queries that have been searched for previously, and then read for the more detailed explanation, which goes way beyond autocomplete.

In the summary section of the description for the patent, we are told about how the patent may address those assumptions:

When anticipating user search requests, responding to the algorithm in this patent can involve certain methods for processing query information. Those include:

  • Receiving query information at a server system, with a portion of a query from a searcher
  • Obtaining a set of predicted queries relevant to the portion of the searcher’s query based on query and data indicative of the searcher relative to previously submitted queries
  • Providing the set of predicted queries to the searcher

The patent also points out additional features involved in the process such as obtaining the predicted queries including ordering the set of predicted queries based upon ranking criteria.

Those ranking criteria may be based upon the data indicative of searcher’s behavior relative to previously submitted queries.

Data about the searcher’s behavior regarding those previously submitted queries may include:

  • Click data
  • Location-specific data
  • Language-specific data
  • Other similar types of data

The patent points out the following as advantages of following the process described in the patent:

A search assistant receives query information from a search requestor, prior to a searcher completely inputting the query.

Information associated with previous user (or users) searches (such as click data associated with search results) is collected. From the query information and the previous search information, a set of predicted queries is produced and provided to the search requestor for presentation.

The patent can be found at:

Autocompletion using previously submitted query data
Inventors: Michael Herscovici, Dan Guez, and Hyung-Jin Kim
Assignee: Google Inc.
US Patent: 9,740,780
Granted: August 22, 2017
Filed: December 1, 2014

Abstract

A computer-implemented method for processing query information includes receiving query information at a server system. The query information includes a portion of a query from a search requestor. The method also includes obtaining a set of predicted queries relevant to the portion of the search requestor query based upon the portion of the query from the search requestor and data indicative of search requestor behavior relative to previously submitted queries. The method also includes providing the set of predicted queries to the search requestor.

Analysis of Ranking and Selection of Search Suggestions Based Upon Previous Query Data

The “Detailed Description” section of this search suggestions patent provides some insightful analysis about search at Google.

Relevance and Backlinks and a Rank Modifying Engine Lead to Ranking For Many Results at Google

This patent points out some of how search works at Google. It tells us that:

  1. The purpose of the process in the patent is to “improve the relevance of results obtained from submitting search queries.”
  2. It describes the ranking of documents for a query as something that can be “performed using traditional techniques for determining an information retrieval (IR) score for indexed documents in view of a given query.” And relevance of a particular document with respect to a query term may be determined by a technique, such as looking at the general level of back-links to a document that contain matches for a search term that may be used to infer a document’s relevance. As the patent tells us:

    In particular, if a document is linked to (e.g., is the target of a hyperlink) by many other relevant documents (e.g., documents that also contain matches for the search terms), it can be inferred that the target document is particularly relevant. This inference can be made because the authors of the pointing documents presumably point, for the most part, to other documents that are relevant to their audience.

  3. We are given more details about some results being even more relevant than ones with backlinks. We are told that:

    If the pointing documents are in turn the targets of links from other relevant documents, they can be considered more relevant, and the first document can be considered particularly relevant because it is the target of relevant (or even highly relevant) documents. Such a technique may be the determinant of a document’s relevance or one of multiple determinants. The technique is exemplified in some systems that treat a link from one web page to another as an indication of quality for the latter page, so that the page with the most such quality indicators is rated higher than others. Appropriate techniques can also be used to identify and eliminate attempts to cast false votes so as to artificially drive up the relevance of a page.

  4. There is another step that could potentially make some results even more relevant that involve what is referred to as a rank modifier engine:

    To further improve such traditional document ranking techniques, the ranking engine can receive an additional signal from a rank modifier engine to assist in determining an appropriate ranking for the documents. The rank modifier engine provides one or more prior models, or one or more measures of relevance for the documents based on one or more prior models, which can be used by the ranking engine to improve the search results’ ranking provided to the user. In general, a prior model represents a background probability of document result selection given the values of multiple selected features, as described further below. The rank modifier engine can perform one or more of the operations described below to generate the one or more prior models, or the one or more measures of relevance based on one or more prior models.

  5. This is a more detailed description of ranking than we normally see at Google. The section above references a Rank Modifier Engine that will be described in more detail further down this post

    Indexing, Scoring, Ranking, and Rank Modifier Engine

    ranking search suggestions

    The information retrieval system from this patent includes a number of different components:

  • Indexing engine
  • Scoring engine
  • Ranking engine
  • Rank modifier engine

The indexing engine can function as described in the section above for the indexing engine.

Scoring Engine

In addition, a scoring engine may provide scores for document results based on many different features including:

  • Content-based features that link a query to document results
  • query-independent features that generally indicate the quality of document results

Content-based features include aspects of document format, such as Query matches to title or anchor text in an HTML (HyperText Markup Language) page.

The query-independent features can include aspects of document cross-referencing, such as a rank of the document or the domain.

Moreover, the particular functions used by the scoring engine can be tuned, to adjust the various feature contributions to the final IR score, using automatic or semi-automatic processes.

Ranking Engine

A ranking engine can produce a ranking of document search results for display to a searcher based on IR scores received from the scoring engine and possibly one or more signals from the rank modifier engine.

A tracking component may be used to record information about individual searcher selections of the search results presented in the ranking. The patent describes how selections may be tracked using javascript or a proxy system or a toolbar plugin:

For example, the tracking component can be embedded JavaScript code included in a web page ranking that identifies user selections (clicks) of individual document results and also identifies when the user returns to the results page, thus indicating the amount of time the user spent viewing the selected document result. In other implementations, the tracking component can be a proxy system through which user selections of the document results are routed, or the tracking component can include pre-installed software at the client (e.g., a toolbar plug-in to the client’s operating system). Other implementations are also possible, such as by using a feature of a web browser that allows a tag/directive to be included in a page, which requests the browser to connect back to the server with message(s) regarding link(s) clicked by the user.

That selection information may also be logged, which could capture for each selection:

  • the query (Q)
  • the document (D)
  • the time (T) on the document
  • the language (L) employed by the user
  • the country (C) where the user is likely located (e.g., based on the server used to access the IR system).

Also other information may be recorded about a searcher’s interactions with presented rankings:

  • Negative information, such as the fact that a document result was presented to a user, but was not clicked
  • Position(s) of click(s) in the user interface
  • IR scores of clicked results
  • IR scores of all results shown before the clicked result
  • Titles and snippets shown to the user before the clicked result
  • The user’s cookie
  • Cookie age
  • IP (Internet Protocol) address
  • User agent of the browser
  • Etc

More information can be recorded (as described in this post below) about building a prior model.

Rank Modifier Engine

Similar information (e.g., IR scores, position, etc.) may be recorded for an entire session, or multiple sessions of a searcher, including possibly recording it for every click that occurs both before and after a current click.

Information that is stored in the result selection logs may be used by the rank modifier engine to generate one or more signals to the ranking engine.

Information stored in the search results selection logs along with the information collected by the tracking component may also be accessible by a search assistant, which is also a component of the information retrieval system.

Along with receiving information from these components, the search assistant could also monitor a user’s entry of a search query.

On receiving a partial search query, the query along with the information (e.g., click data) from the tracking component and the results selection log(s) may be used to predict a searcher’s contemplated complete query.

Based on this information, predictions may be ordered according to one or more ranking criteria before being presented to assist the user in completing the query.

Presentation of a Search Suggestion

As a searcher enters a search query, the searcher’s input is monitored.

Before the searcher signals that they have completed entering the search query, a portion of the query is sent to the search engine.

Also, data such as click data (or other types of previously collected information) may also be sent with the query portion.

The portion of the query sent may be:

  • A few characters
  • A search term
  • More than one search term
  • Any other combination of characters and terms

The search engine receives the partial query and the data (e.g., click data) for processing and makes predictions) as to the searcher’s contemplated complete query.

Relevant information may be retrieved for processing with the received partial query to produce search suggestions predictions.

Predictions may be ordered according to one or more ranking criteria.

So, queries that have been submitted at a higher frequency may be ordered before queries submitted at lower frequencies.

The search engine may also use various types of information for ranking and ordering predicted queries as search suggestions.

Information about previously entered search queries may be used to make ordered predictions.

Previous queries may include search queries associated with the same user, another user, or from a community of users.

If one of the predicted queries is what the searcher intended as the desired query, the searcher may select that predicted query and proceed without having to finish entering the desired query.

Alternatively, if the predicted queries do not reflect what the searcher had in mind, then the searcher can continue entering the desired search query, which could trigger one or more other sets of search suggestions.

Ranking User Submitted Previous Queries as Search Suggestions

The patent tells us that a few different processes may be used in ranking and ordering predicted search queries:

  • Predicted search queries may be ordered in accordance with a frequency of submission by a community of users
  • Time constraints may also be used with search queries ordered in accordance with the last time/date value that the query was submitted
  • Personalization information or community information may be used such as information about subjects, concepts or categories of information that are of interest to the user (from prior search or browsing information)
  • Personalization may also be from a group that the searcher is associated with or belongs to (a member or an employee.)
  • According to a first ranking criteria, such as predefined popularity criteria, and then possibly reordered if any of the predicted search queries match the user personalization information of the user, to place the matching predicted search queries at or closer to the top of the ordered set of predicted search queries
  • Information provided by the tracking component and the result selection log(s) might be used for ranking and ordering the predicted search queries. (click data, language-specific, and country-specific data.)
  • Processed click data (e.g., aggregated click data for a given query) could be used for ranking and ordering predicted search queries – or each query a score may be calculated by summing click data (e.g., weighted clicks, etc.) on documents associated with the query, and predicted queries may be ordered based upon the score (e.g., higher values representing better)

An Information Model Based On Previously Submitted Query Data to Obtain Search Suggestions Predictions

This model can be used to predict what query data might satisfy a searcher the most by looking at long click information. A timer can be used to track how long a user views or “dwells” on a document.

The amount of time is referred to as “click data”.

A longer time spent dwelling on a document, would be termed a “long click”, and can indicate that a user found the document to be relevant for their query.

A brief period viewing a document would be termed a “short click”, and can be interpreted as a lack of document relevance.

Click data is a count of each click type (e.g., long, medium, short) for a particular query and document combination.

This aggregated click data from model queries for a given document can be used to create a quality of result statistic for that document to enhance a ranking of that document.

Quality of result statistic can be a weighted average of the count of long clicks for a given document and query.

This description from the patent tells us about how click data might be stored in tuples:

A search engine (e.g., the search engine) or other processes may create a record in the model for documents that are selected by users in response to a query or a partial query. Each record within the model (herein referred to as a tuple: ) is at least a combination of a query submitted by users, a document reference selected by users in response to that query, and an aggregation of click data for all users that select the document reference in response to the query. The aggregate click data can be viewed as an indication of document relevance. In various implementations, model data can be location-specific (e.g. country, state, etc) or language-specific. For example, a country-specific tuple would include the country from where the user query originated from in whereas a language-specific tuple would include the language of the user query. Other extensions of model data are possible.

The model may also include Post-click behavior that has been tracked by the tracking component.

This patent does include a lot of information about how Google might use click tracking data when ranking search suggestion predictions. It tells us about the data that could be collected about clicks:

The information gathered for each click can include:

(1) the query (Q) the user entered,
(2) the document result (D) the user clicked on,
(3) the time (T) on the document,
(4) the interface language (L) (which can be given by the user),
(5) the country (C) of the user (which can be identified by the host that they use, such as www-store-co-uk to indicate the United Kingdom), and
(6) additional aspects of the user and session.

The time (T) can be measured as the time between the initial click through to the document result until the time the user comes back to the main page and clicks on another document result. Moreover, an assessment can be made about the time (T) regarding whether this time indicates a longer view of the document result or a shorter view of the document result, since longer views are generally indicative of quality for the clicked through result. This assessment about the time (T) can further be made in conjunction with various weighting techniques.

Beyond Long Clicks

We are also told that document views from the selections can be weighted based on viewing length information to produce weighted views of the document result.

So, rather than simply distinguishing long clicks from short clicks, a wider range of click through viewing times can be included in the assessment of result quality, where longer viewing times in the range are given more weight than shorter viewing times.

Predicted Search Suggestions

Google will sometimes display search suggestions using autocomplete and also based upon user data from previous queries from a searcher’s previous search history, or the history of someone whom the searcher may be associated with, such as a fellow member of an organization or a co-worker.

While results related to those previous queries were ranked based upon such things as relevance and backlinks, the search suggestions may include results that searchers spent long clicks upon, including long times viewing.

So pursuant to this patent, predictions about search suggestions chosen using autocomplete may best meet a searcher’s informational needs by being searches that include results remembered as resulting in long clicks and long viewing times.


Copyright © 2020 SEO by the Sea ⚓. This Feed is for personal non-commercial use only. If you are not reading this material in your news aggregator, the site you are looking at may be guilty of copyright infringement. Please contact SEO by the Sea, so we can take appropriate action immediately.
Plugin by Taragana

Search Suggestions from Previously Submitted Searcher Queries was originally posted by Video And Blog Marketing

Search Suggestions from Previously Submitted Searcher Queries

I came across an interesting Search Engine Land post last week. It inspired me to search and see if I could find a patent that might be related to it from Google:

Google is suggesting searches based on users’ recent activity

I tried to reproduce the search suggestions that were being shown to the author of the Search Engine Land article, but Google would not return those to me. Google may be experimenting with a limited number of searchers rather than showing those results to everyone. I did find a patent discussing search suggestions that were similar.

When Google shows a search suggestion about something you may have searched for in the past, that predicted suggestion is likely related to a patent I’ve written about before, Autocompletion using previously submitted query data.

I wrote about that patent being updated in a continuation patent, but hadn’t provided much in the way of details about how it works at: How Google Predicts Autocomplete Query Suggestions is Updated.

There are some interesting parts about how search suggestions are identified and ranked, which inspired me to write this post.

Search Suggestions Based on Previously Submitted Query Data

The description of this patent starts off by telling us that it is about: “using previously submitted query data to anticipate a user’s search request.”

That pinpoints that Google has a long memory, and it remembers a lot about what someone might search for.

This patent description also includes a lot of the assumptions that search engineers make about searchers (often an interesting reason to read through patents). Here are some from this patent that are worth thinking about:

Internet search engines aim to identify documents or other items that are relevant to a user’s needs and to present the documents or items in a manner that is most useful to the user. Such activity often involves a fair amount of mind-reading–inferring from various clues what the user wants. Certain clues may be user-specific. For example, the knowledge that a user is making a request from a mobile device, and knowledge of the location of the device, can result in much better search results for such a user.

Clues about a user’s needs may also be more general. For example, search results can have elevated importance, or inferred relevance, if a number of other search results link to them. If the linking results are themselves highly relevant, then the linked-to results may have particularly high relevance. Such an approach to determining relevance may be premised on the assumption that, if authors of web pages felt that another web site was relevant enough to be linked to, then web searchers would also find the site to be particularly relevant. In short, the web authors “vote up” the relevance of the sites.

Other various inputs may be used instead of, or in addition to, such techniques for determining and ranking search results. For example, user reactions to particular search results or search result lists may be gauged, so that results on which users often click will receive a higher ranking. The general assumption under such an approach is that searching users are often the best judges of relevance, so that if they select a particular search result, it is likely to be relevant, or at least more relevant than the presented alternatives.

A Summary of the Search Suggestions Process Based on Previous Submitted Queries

Like most patents, the Description for this one starts out with a summary section that provides an overview of how the process defined in the patent works. It is followed by a “Detailed Description” section that goes into more depth and provides details about how search at Google works, and how specific aspects of search at Google power this search suggestion process. So read about how search suggestions might be provided based upon user queries that have been searched for previously, and then read for the more detailed explanation, which goes way beyond autocomplete.

In the summary section of the description for the patent, we are told about how the patent may address those assumptions:

When anticipating user search requests, responding to the algorithm in this patent can involve certain methods for processing query information. Those include:

  • Receiving query information at a server system, with a portion of a query from a searcher
  • Obtaining a set of predicted queries relevant to the portion of the searcher’s query based on query and data indicative of the searcher relative to previously submitted queries
  • Providing the set of predicted queries to the searcher

The patent also points out additional features involved in the process such as obtaining the predicted queries including ordering the set of predicted queries based upon ranking criteria.

Those ranking criteria may be based upon the data indicative of searcher’s behavior relative to previously submitted queries.

Data about the searcher’s behavior regarding those previously submitted queries may include:

  • Click data
  • Location-specific data
  • Language-specific data
  • Other similar types of data

The patent points out the following as advantages of following the process described in the patent:

A search assistant receives query information from a search requestor, prior to a searcher completely inputting the query.

Information associated with previous user (or users) searches (such as click data associated with search results) is collected. From the query information and the previous search information, a set of predicted queries is produced and provided to the search requestor for presentation.

The patent can be found at:

Autocompletion using previously submitted query data
Inventors: Michael Herscovici, Dan Guez, and Hyung-Jin Kim
Assignee: Google Inc.
US Patent: 9,740,780
Granted: August 22, 2017
Filed: December 1, 2014

Abstract

A computer-implemented method for processing query information includes receiving query information at a server system. The query information includes a portion of a query from a search requestor. The method also includes obtaining a set of predicted queries relevant to the portion of the search requestor query based upon the portion of the query from the search requestor and data indicative of search requestor behavior relative to previously submitted queries. The method also includes providing the set of predicted queries to the search requestor.

Analysis of Ranking and Selection of Search Suggestions Based Upon Previous Query Data

The “Detailed Description” section of this search suggestions patent provides some insightful analysis about search at Google.

Relevance and Backlinks and a Rank Modifying Engine Lead to Ranking For Many Results at Google

This patent points out some of how search works at Google. It tells us that:

  1. The purpose of the process in the patent is to “improve the relevance of results obtained from submitting search queries.”
  2. It describes the ranking of documents for a query as something that can be “performed using traditional techniques for determining an information retrieval (IR) score for indexed documents in view of a given query.” And relevance of a particular document with respect to a query term may be determined by a technique, such as looking at the general level of back-links to a document that contain matches for a search term that may be used to infer a document’s relevance. As the patent tells us:

    In particular, if a document is linked to (e.g., is the target of a hyperlink) by many other relevant documents (e.g., documents that also contain matches for the search terms), it can be inferred that the target document is particularly relevant. This inference can be made because the authors of the pointing documents presumably point, for the most part, to other documents that are relevant to their audience.

  3. We are given more details about some results being even more relevant than ones with backlinks. We are told that:

    If the pointing documents are in turn the targets of links from other relevant documents, they can be considered more relevant, and the first document can be considered particularly relevant because it is the target of relevant (or even highly relevant) documents. Such a technique may be the determinant of a document’s relevance or one of multiple determinants. The technique is exemplified in some systems that treat a link from one web page to another as an indication of quality for the latter page, so that the page with the most such quality indicators is rated higher than others. Appropriate techniques can also be used to identify and eliminate attempts to cast false votes so as to artificially drive up the relevance of a page.

  4. There is another step that could potentially make some results even more relevant that involve what is referred to as a rank modifier engine:

    To further improve such traditional document ranking techniques, the ranking engine can receive an additional signal from a rank modifier engine to assist in determining an appropriate ranking for the documents. The rank modifier engine provides one or more prior models, or one or more measures of relevance for the documents based on one or more prior models, which can be used by the ranking engine to improve the search results’ ranking provided to the user. In general, a prior model represents a background probability of document result selection given the values of multiple selected features, as described further below. The rank modifier engine can perform one or more of the operations described below to generate the one or more prior models, or the one or more measures of relevance based on one or more prior models.

  5. This is a more detailed description of ranking than we normally see at Google. The section above references a Rank Modifier Engine that will be described in more detail further down this post

    Indexing, Scoring, Ranking, and Rank Modifier Engine

    ranking search suggestions

    The information retrieval system from this patent includes a number of different components:

  • Indexing engine
  • Scoring engine
  • Ranking engine
  • Rank modifier engine

The indexing engine can function as described in the section above for the indexing engine.

Scoring Engine

In addition, a scoring engine may provide scores for document results based on many different features including:

  • Content-based features that link a query to document results
  • query-independent features that generally indicate the quality of document results

Content-based features include aspects of document format, such as Query matches to title or anchor text in an HTML (HyperText Markup Language) page.

The query-independent features can include aspects of document cross-referencing, such as a rank of the document or the domain.

Moreover, the particular functions used by the scoring engine can be tuned, to adjust the various feature contributions to the final IR score, using automatic or semi-automatic processes.

Ranking Engine

A ranking engine can produce a ranking of document search results for display to a searcher based on IR scores received from the scoring engine and possibly one or more signals from the rank modifier engine.

A tracking component may be used to record information about individual searcher selections of the search results presented in the ranking. The patent describes how selections may be tracked using javascript or a proxy system or a toolbar plugin:

For example, the tracking component can be embedded JavaScript code included in a web page ranking that identifies user selections (clicks) of individual document results and also identifies when the user returns to the results page, thus indicating the amount of time the user spent viewing the selected document result. In other implementations, the tracking component can be a proxy system through which user selections of the document results are routed, or the tracking component can include pre-installed software at the client (e.g., a toolbar plug-in to the client’s operating system). Other implementations are also possible, such as by using a feature of a web browser that allows a tag/directive to be included in a page, which requests the browser to connect back to the server with message(s) regarding link(s) clicked by the user.

That selection information may also be logged, which could capture for each selection:

  • the query (Q)
  • the document (D)
  • the time (T) on the document
  • the language (L) employed by the user
  • the country (C) where the user is likely located (e.g., based on the server used to access the IR system).

Also other information may be recorded about a searcher’s interactions with presented rankings:

  • Negative information, such as the fact that a document result was presented to a user, but was not clicked
  • Position(s) of click(s) in the user interface
  • IR scores of clicked results
  • IR scores of all results shown before the clicked result
  • Titles and snippets shown to the user before the clicked result
  • The user’s cookie
  • Cookie age
  • IP (Internet Protocol) address
  • User agent of the browser
  • Etc

More information can be recorded (as described in this post below) about building a prior model.

Rank Modifier Engine

Similar information (e.g., IR scores, position, etc.) may be recorded for an entire session, or multiple sessions of a searcher, including possibly recording it for every click that occurs both before and after a current click.

Information that is stored in the result selection logs may be used by the rank modifier engine to generate one or more signals to the ranking engine.

Information stored in the search results selection logs along with the information collected by the tracking component may also be accessible by a search assistant, which is also a component of the information retrieval system.

Along with receiving information from these components, the search assistant could also monitor a user’s entry of a search query.

On receiving a partial search query, the query along with the information (e.g., click data) from the tracking component and the results selection log(s) may be used to predict a searcher’s contemplated complete query.

Based on this information, predictions may be ordered according to one or more ranking criteria before being presented to assist the user in completing the query.

Presentation of a Search Suggestion

As a searcher enters a search query, the searcher’s input is monitored.

Before the searcher signals that they have completed entering the search query, a portion of the query is sent to the search engine.

Also, data such as click data (or other types of previously collected information) may also be sent with the query portion.

The portion of the query sent may be:

  • A few characters
  • A search term
  • More than one search term
  • Any other combination of characters and terms

The search engine receives the partial query and the data (e.g., click data) for processing and makes predictions) as to the searcher’s contemplated complete query.

Relevant information may be retrieved for processing with the received partial query to produce search suggestions predictions.

Predictions may be ordered according to one or more ranking criteria.

So, queries that have been submitted at a higher frequency may be ordered before queries submitted at lower frequencies.

The search engine may also use various types of information for ranking and ordering predicted queries as search suggestions.

Information about previously entered search queries may be used to make ordered predictions.

Previous queries may include search queries associated with the same user, another user, or from a community of users.

If one of the predicted queries is what the searcher intended as the desired query, the searcher may select that predicted query and proceed without having to finish entering the desired query.

Alternatively, if the predicted queries do not reflect what the searcher had in mind, then the searcher can continue entering the desired search query, which could trigger one or more other sets of search suggestions.

Ranking User Submitted Previous Queries as Search Suggestions

The patent tells us that a few different processes may be used in ranking and ordering predicted search queries:

  • Predicted search queries may be ordered in accordance with a frequency of submission by a community of users
  • Time constraints may also be used with search queries ordered in accordance with the last time/date value that the query was submitted
  • Personalization information or community information may be used such as information about subjects, concepts or categories of information that are of interest to the user (from prior search or browsing information)
  • Personalization may also be from a group that the searcher is associated with or belongs to (a member or an employee.)
  • According to a first ranking criteria, such as predefined popularity criteria, and then possibly reordered if any of the predicted search queries match the user personalization information of the user, to place the matching predicted search queries at or closer to the top of the ordered set of predicted search queries
  • Information provided by the tracking component and the result selection log(s) might be used for ranking and ordering the predicted search queries. (click data, language-specific, and country-specific data.)
  • Processed click data (e.g., aggregated click data for a given query) could be used for ranking and ordering predicted search queries – or each query a score may be calculated by summing click data (e.g., weighted clicks, etc.) on documents associated with the query, and predicted queries may be ordered based upon the score (e.g., higher values representing better)

An Information Model Based On Previously Submitted Query Data to Obtain Search Suggestions Predictions

This model can be used to predict what query data might satisfy a searcher the most by looking at long click information. A timer can be used to track how long a user views or “dwells” on a document.

The amount of time is referred to as “click data”.

A longer time spent dwelling on a document, would be termed a “long click”, and can indicate that a user found the document to be relevant for their query.

A brief period viewing a document would be termed a “short click”, and can be interpreted as a lack of document relevance.

Click data is a count of each click type (e.g., long, medium, short) for a particular query and document combination.

This aggregated click data from model queries for a given document can be used to create a quality of result statistic for that document to enhance a ranking of that document.

Quality of result statistic can be a weighted average of the count of long clicks for a given document and query.

This description from the patent tells us about how click data might be stored in tuples:

A search engine (e.g., the search engine) or other processes may create a record in the model for documents that are selected by users in response to a query or a partial query. Each record within the model (herein referred to as a tuple: ) is at least a combination of a query submitted by users, a document reference selected by users in response to that query, and an aggregation of click data for all users that select the document reference in response to the query. The aggregate click data can be viewed as an indication of document relevance. In various implementations, model data can be location-specific (e.g. country, state, etc) or language-specific. For example, a country-specific tuple would include the country from where the user query originated from in whereas a language-specific tuple would include the language of the user query. Other extensions of model data are possible.

The model may also include Post-click behavior that has been tracked by the tracking component.

This patent does include a lot of information about how Google might use click tracking data when ranking search suggestion predictions. It tells us about the data that could be collected about clicks:

The information gathered for each click can include:

(1) the query (Q) the user entered,
(2) the document result (D) the user clicked on,
(3) the time (T) on the document,
(4) the interface language (L) (which can be given by the user),
(5) the country (C) of the user (which can be identified by the host that they use, such as www-store-co-uk to indicate the United Kingdom), and
(6) additional aspects of the user and session.

The time (T) can be measured as the time between the initial click through to the document result until the time the user comes back to the main page and clicks on another document result. Moreover, an assessment can be made about the time (T) regarding whether this time indicates a longer view of the document result or a shorter view of the document result, since longer views are generally indicative of quality for the clicked through result. This assessment about the time (T) can further be made in conjunction with various weighting techniques.

Beyond Long Clicks

We are also told that document views from the selections can be weighted based on viewing length information to produce weighted views of the document result.

So, rather than simply distinguishing long clicks from short clicks, a wider range of click through viewing times can be included in the assessment of result quality, where longer viewing times in the range are given more weight than shorter viewing times.

Predicted Search Suggestions

Google will sometimes display search suggestions using autocomplete and also based upon user data from previous queries from a searcher’s previous search history, or the history of someone whom the searcher may be associated with, such as a fellow member of an organization or a co-worker.

While results related to those previous queries were ranked based upon such things as relevance and backlinks, the search suggestions may include results that searchers spent long clicks upon, including long times viewing.

So pursuant to this patent, predictions about search suggestions chosen using autocomplete may best meet a searcher’s informational needs by being searches that include results remembered as resulting in long clicks and long viewing times.


Copyright © 2020 SEO by the Sea ⚓. This Feed is for personal non-commercial use only. If you are not reading this material in your news aggregator, the site you are looking at may be guilty of copyright infringement. Please contact SEO by the Sea, so we can take appropriate action immediately.
Plugin by Taragana

Search Suggestions from Previously Submitted Searcher Queries was originally posted by Video And Blog Marketing

The Best Way to Find 404 Pages and How to Fix Them

Last updated on

The Best way to Find 404 Pages and How to Fix them_Cover Photo

There are a lot of tools available in the market today that enable webmasters and SEOs to find broken pages inside their website. But not one of them can show you the most important broken pages that users still enter. This means that the broken page where users go to, if not fixed immediately, will make you lose valuable traffic in the long run. 

However, there is an easier and faster way to find these important broken pages and be able to fix them immediately. The surprising part about this is possible through a tool that ALL webmasters and SEOs use (and the tool is absolutely free). So, what’s the best way to find important 404 pages?

Use Google Analytics to Find 404 Pages

Simple enough, Google Analytics can show you the broken pages that are garnering traffic from your users. And it’s easy enough that anybody can do it as long as they have access to the website’s Google Analytics account. Here’s how to do it:

  • Go to your Google Analytics Account
  • Click the Behavior drop-down Button > Click on the Site Content drop-down button > Then click All Pages.

Google Analytics Behavior Site Content screenshot

  • In the list showing your top pages, choose Page Title as the primary dimension

All Pages Google Analytics Screenshot

  • Use the search bar to search for 404 pages. They commonly have “404” or “Page Not Found” as their Page title

Search bar for all pages in behavior Google Analytics

  • After searching, you’ll be able to click on the result that contains all the URLs that have a 404 status code

404 Pages in GOogle Analytics Screenshot

  • Click on the result, compile the list of broken pages

list of 404s in Google Analytics

  • Fix the pages you found.

How to Fix 404 Pages

Now that you’ve found the most important 404 pages, it’s time for you to fix them to not lose all this valuable traffic. You can do it in two ways: If you have website development knowledge, fix it immediately, or contact your developer and ask them to fix it as soon as possible. Once you fix it, the page will serve its purpose, and hopefully, be able to give the user what they’re looking for.

The second way is to redirect the broken page to a relevant but working page. This enables you to save the user from bouncing but this does come with the risk of them still bouncing if the page you’re redirecting them to does not give them what they’re looking for.  

Key Takeaway

404 Errors are one of the most common problems websites experience. It doesn’t matter if its a newly-established website or a well-known, high-traffic website, all of them are susceptible to 404 errors. The easiest way to resolve these issues would be to redirect them but there will also be times when redirection is bad for SEO. So, keep this in mind.

Now that you know the best way to finding 404 errors and how to fix them, you’ll now be able to retain the traffic that you were losing while the 404 errors existed. Do you have any other way to find 404 errors inside your website? Comment them down below!

The Best Way to Find 404 Pages and How to Fix Them was originally posted by Video And Blog Marketing

A week into Google’s May 2020 Core Update

Last updated on

A Week Into Google's May 2020 Update_Cover Photo

Before the May 2020 Broad Core Update was released, there have been talks about it among the SEO community. “Surely Google would not release an update during this time,” but lo and behold they have announced it on the 5th of May that left some of us dumbfounded on the results that they have received. The announcement has kept us on our toes as most updates go, they usually roll out during the first or second week of implementation.

With this in mind, I have constantly monitored my sites and my clients as well to see if any of them came out strong or if they have been hurt by the update.

Winners and Losers in the May 2020 Update

Before we get into the personal side of things, let’s look at the volatility of the search results since the update.

Cognitive SEO Signals

Accuranker Google Grump

accuranker google grump

SEMrush Sensor

semrush serp volatilty

As you can see, the volatility of the search results is wild. Fluctuation is prevalent among sites and this has enjoined webmasters to stay vigilant on how to maintain steady traction or to recover from their losses. This is the perfect time to work on website fixes to see if you can hit that sweet spot that the algorithm will honor in the weeks to come. Since the update is still fresh and will not be fully rolled out until its 2nd week, this can buy you time to see what works and what doesn’t.

Anyway, I have told you that I have seen the effect on some of my client’s sites and I have found that winners are recipe sites, automotive industries, and e-commerce sites. For the health and news sector, there has been a dramatic drop. Before we discuss the winners and losers from this update, I would like to throw it out there that if your site has been affected by the update, it does not mean that your industry is targeted by the algorithm. This simply implies that updates have been magnifying their evaluation of sites continuously optimizing for search intent and relevance for search queries.

Moving on, here is what the data dictates for the sites I am monitoring:

Winners

ecommerce site

This e-commerce client has been enjoying a ton of traffic and is one of the testaments that this lockdown is viable for digital marketing efforts’ contribution to businesses. It is now strong in terms of online visibility with results going as far as a 100% increase in traffic but with the core update, it is showing even more signs of an increase.

recipe site core update

Recipe sites are also one of the winners during this core update and the site I am handling that mainly deals with recipes for their content has been positively impacted by the update. Compared to the previous period, there has been a 35.75% increase in users and 29.14% in sessions. This has solidified its place as a winner from this core update. However, there are some webmasters that are not so lucky, with some of them saying that their rich snippets have been removed resulting in a drastic drop in traffic.

toyota

The automotive industry has also shown significant results when it comes to traffic increase. Here in the Philippines, modes of transportation are at a standstill and this event affects user behavior in terms of looking for alternative ways to get around their location. Be that as it may, the update has been favorable to the sites for this industry. As you can see from the noticeable spike in traffic on May 5th, this site has also enjoyed a favorable appraisal from the algorithm broad core update.

Losers

We are moving over to the unfortunate few who have been hit by the update. The drop in traffic shown on these sites may be attributed to the core update, ranging from health sites to industry news sites. Let’s look at these losers from the May 2020 core update.

health site - may update

For a health site, it seems that it has been drastically affected by a significant amount of traffic shaved off during the first week of May. Traffic started slipping on May 5 and has not recovered since then. This is in comparison to the previous period wherein the site has experienced a steady stream of users going into the site. With the search algorithm clamping down on content pertaining to the pandemic, this may be another way for Google to evaluate the quality of health sites.

industry news core update

An industry news site is also affected by the update. The site maintained its strength in traffic up until May 9 when it started dipping and hasn’t increased since. Although the lost traffic is just a little over a thousand, it is still a significant loss for this site. 14% of traffic slipped away compared to last week.

Key Takeaway

The Core Update affects health sectors the most as Google continues to pay attention to news concerning the pandemic. It seems this is an urgent call to uphold the standard when it comes to search queries and remains faithful to giving the right answers to it. If people are looking for scientific explanations for information dissemination on the coronavirus, chances are, Google will honor content that is highly-reputable for this purpose. We are living in a time when misinformation can cost a life, so it is only admirable that the algorithm is constantly evolving to help champion this cause.

So far, have you seen the effects of the update on your site? Do let me know in the comments below.

A week into Google’s May 2020 Core Update was originally posted by Video And Blog Marketing

Search Suggestions from Previously Submitted Searcher Queries

I came across an interesting Search Engine Land post last week. It inspired me to search and see if I could find a patent that might be related to it from Google:

Google is suggesting searches based on users’ recent activity

I tried to reproduce the search suggestions that were being shown to the author of the Search Engine Land article, but Google would not return those to me. Google may be experimenting with a limited number of searchers rather than showing those results to everyone. I did find a patent discussing search suggestions that were similar.

When Google shows a search suggestion about something you may have searched for in the past, that predicted suggestion is likely related to a patent I’ve written about before, Autocompletion using previously submitted query data.

I wrote about that patent being updated in a continuation patent, but hadn’t provided much in the way of details about how it works at: How Google Predicts Autocomplete Query Suggestions is Updated.

There are some interesting parts about how search suggestions are identified and ranked, which inspired me to write this post.

Search Suggestions Based on Previously Submitted Query Data

The description of this patent starts off by telling us that it is about: “using previously submitted query data to anticipate a user’s search request.”

That pinpoints that Google has a long memory, and it remembers a lot about what someone might search for.

This patent description also includes a lot of the assumptions that search engineers make about searchers (often an interesting reason to read through patents). Here are some from this patent that are worth thinking about:

Internet search engines aim to identify documents or other items that are relevant to a user’s needs and to present the documents or items in a manner that is most useful to the user. Such activity often involves a fair amount of mind-reading–inferring from various clues what the user wants. Certain clues may be user-specific. For example, the knowledge that a user is making a request from a mobile device, and knowledge of the location of the device, can result in much better search results for such a user.

Clues about a user’s needs may also be more general. For example, search results can have elevated importance, or inferred relevance, if a number of other search results link to them. If the linking results are themselves highly relevant, then the linked-to results may have particularly high relevance. Such an approach to determining relevance may be premised on the assumption that, if authors of web pages felt that another web site was relevant enough to be linked to, then web searchers would also find the site to be particularly relevant. In short, the web authors “vote up” the relevance of the sites.

Other various inputs may be used instead of, or in addition to, such techniques for determining and ranking search results. For example, user reactions to particular search results or search result lists may be gauged, so that results on which users often click will receive a higher ranking. The general assumption under such an approach is that searching users are often the best judges of relevance, so that if they select a particular search result, it is likely to be relevant, or at least more relevant than the presented alternatives.

A Summary of the Search Suggestions Process Based on Previous Submitted Queries

Like most patents, the Description for this one starts out with a summary section that provides an overview of how the process defined in the patent works. It is followed by a “Detailed Description” section that goes into more depth and provides details about how search at Google works, and how specific aspects of search at Google power this search suggestion process. So read about how search suggestions might be provided based upon user queries that have been searched for previously, and then read for the more detailed explanation, which goes way beyond autocomplete.

In the summary section of the description for the patent, we are told about how the patent may address those assumptions:

When anticipating user search requests, responding to the algorithm in this patent can involve certain methods for processing query information. Those include:

  • Receiving query information at a server system, with a portion of a query from a searcher
  • Obtaining a set of predicted queries relevant to the portion of the searcher’s query based on query and data indicative of the searcher relative to previously submitted queries
  • Providing the set of predicted queries to the searcher

The patent also points out additional features involved in the process such as obtaining the predicted queries including ordering the set of predicted queries based upon ranking criteria.

Those ranking criteria may be based upon the data indicative of searcher’s behavior relative to previously submitted queries.

Data about the searcher’s behavior regarding those previously submitted queries may include:

  • Click data
  • Location-specific data
  • Language-specific data
  • Other similar types of data

The patent points out the following as advantages of following the process described in the patent:

A search assistant receives query information from a search requestor, prior to a searcher completely inputting the query.

Information associated with previous user (or users) searches (such as click data associated with search results) is collected. From the query information and the previous search information, a set of predicted queries is produced and provided to the search requestor for presentation.

The patent can be found at:

Autocompletion using previously submitted query data
Inventors: Michael Herscovici, Dan Guez, and Hyung-Jin Kim
Assignee: Google Inc.
US Patent: 9,740,780
Granted: August 22, 2017
Filed: December 1, 2014

Abstract

A computer-implemented method for processing query information includes receiving query information at a server system. The query information includes a portion of a query from a search requestor. The method also includes obtaining a set of predicted queries relevant to the portion of the search requestor query based upon the portion of the query from the search requestor and data indicative of search requestor behavior relative to previously submitted queries. The method also includes providing the set of predicted queries to the search requestor.

Analysis of Ranking and Selection of Search Suggestions Based Upon Previous Query Data

The “Detailed Description” section of this search suggestions patent provides some insightful analysis about search at Google.

Relevance and Backlinks and a Rank Modifying Engine Lead to Ranking For Many Results at Google

This patent points out some of how search works at Google. It tells us that:

  1. The purpose of the process in the patent is to “improve the relevance of results obtained from submitting search queries.”
  2. It describes the ranking of documents for a query as something that can be “performed using traditional techniques for determining an information retrieval (IR) score for indexed documents in view of a given query.” And relevance of a particular document with respect to a query term may be determined by a technique, such as looking at the general level of back-links to a document that contain matches for a search term that may be used to infer a document’s relevance. As the patent tells us:

    In particular, if a document is linked to (e.g., is the target of a hyperlink) by many other relevant documents (e.g., documents that also contain matches for the search terms), it can be inferred that the target document is particularly relevant. This inference can be made because the authors of the pointing documents presumably point, for the most part, to other documents that are relevant to their audience.

  3. We are given more details about some results being even more relevant than ones with backlinks. We are told that:

    If the pointing documents are in turn the targets of links from other relevant documents, they can be considered more relevant, and the first document can be considered particularly relevant because it is the target of relevant (or even highly relevant) documents. Such a technique may be the determinant of a document’s relevance or one of multiple determinants. The technique is exemplified in some systems that treat a link from one web page to another as an indication of quality for the latter page, so that the page with the most such quality indicators is rated higher than others. Appropriate techniques can also be used to identify and eliminate attempts to cast false votes so as to artificially drive up the relevance of a page.

  4. There is another step that could potentially make some results even more relevant that involve what is referred to as a rank modifier engine:

    To further improve such traditional document ranking techniques, the ranking engine can receive an additional signal from a rank modifier engine to assist in determining an appropriate ranking for the documents. The rank modifier engine provides one or more prior models, or one or more measures of relevance for the documents based on one or more prior models, which can be used by the ranking engine to improve the search results’ ranking provided to the user. In general, a prior model represents a background probability of document result selection given the values of multiple selected features, as described further below. The rank modifier engine can perform one or more of the operations described below to generate the one or more prior models, or the one or more measures of relevance based on one or more prior models.

  5. This is a more detailed description of ranking than we normally see at Google. The section above references a Rank Modifier Engine that will be described in more detail further down this post

    Indexing, Scoring, Ranking, and Rank Modifier Engine

    ranking search suggestions

    The information retrieval system from this patent includes a number of different components:

  • Indexing engine
  • Scoring engine
  • Ranking engine
  • Rank modifier engine

The indexing engine can function as described in the section above for the indexing engine.

Scoring Engine

In addition, a scoring engine may provide scores for document results based on many different features including:

  • Content-based features that link a query to document results
  • query-independent features that generally indicate the quality of document results

Content-based features include aspects of document format, such as Query matches to title or anchor text in an HTML (HyperText Markup Language) page.

The query-independent features can include aspects of document cross-referencing, such as a rank of the document or the domain.

Moreover, the particular functions used by the scoring engine can be tuned, to adjust the various feature contributions to the final IR score, using automatic or semi-automatic processes.

Ranking Engine

A ranking engine can produce a ranking of document search results for display to a searcher based on IR scores received from the scoring engine and possibly one or more signals from the rank modifier engine.

A tracking component may be used to record information about individual searcher selections of the search results presented in the ranking. The patent describes how selections may be tracked using javascript or a proxy system or a toolbar plugin:

For example, the tracking component can be embedded JavaScript code included in a web page ranking that identifies user selections (clicks) of individual document results and also identifies when the user returns to the results page, thus indicating the amount of time the user spent viewing the selected document result. In other implementations, the tracking component can be a proxy system through which user selections of the document results are routed, or the tracking component can include pre-installed software at the client (e.g., a toolbar plug-in to the client’s operating system). Other implementations are also possible, such as by using a feature of a web browser that allows a tag/directive to be included in a page, which requests the browser to connect back to the server with message(s) regarding link(s) clicked by the user.

That selection information may also be logged, which could capture for each selection:

  • the query (Q)
  • the document (D)
  • the time (T) on the document
  • the language (L) employed by the user
  • the country (C) where the user is likely located (e.g., based on the server used to access the IR system).

Also other information may be recorded about a searcher’s interactions with presented rankings:

  • Negative information, such as the fact that a document result was presented to a user, but was not clicked
  • Position(s) of click(s) in the user interface
  • IR scores of clicked results
  • IR scores of all results shown before the clicked result
  • Titles and snippets shown to the user before the clicked result
  • The user’s cookie
  • Cookie age
  • IP (Internet Protocol) address
  • User agent of the browser
  • Etc

More information can be recorded (as described in this post below) about building a prior model.

Rank Modifier Engine

Similar information (e.g., IR scores, position, etc.) may be recorded for an entire session, or multiple sessions of a searcher, including possibly recording it for every click that occurs both before and after a current click.

Information that is stored in the result selection logs may be used by the rank modifier engine to generate one or more signals to the ranking engine.

Information stored in the search results selection logs along with the information collected by the tracking component may also be accessible by a search assistant, which is also a component of the information retrieval system.

Along with receiving information from these components, the search assistant could also monitor a user’s entry of a search query.

On receiving a partial search query, the query along with the information (e.g., click data) from the tracking component and the results selection log(s) may be used to predict a searcher’s contemplated complete query.

Based on this information, predictions may be ordered according to one or more ranking criteria before being presented to assist the user in completing the query.

Presentation of a Search Suggestion

As a searcher enters a search query, the searcher’s input is monitored.

Before the searcher signals that they have completed entering the search query, a portion of the query is sent to the search engine.

Also, data such as click data (or other types of previously collected information) may also be sent with the query portion.

The portion of the query sent may be:

  • A few characters
  • A search term
  • More than one search term
  • Any other combination of characters and terms

The search engine receives the partial query and the data (e.g., click data) for processing and makes predictions) as to the searcher’s contemplated complete query.

Relevant information may be retrieved for processing with the received partial query to produce search suggestions predictions.

Predictions may be ordered according to one or more ranking criteria.

So, queries that have been submitted at a higher frequency may be ordered before queries submitted at lower frequencies.

The search engine may also use various types of information for ranking and ordering predicted queries as search suggestions.

Information about previously entered search queries may be used to make ordered predictions.

Previous queries may include search queries associated with the same user, another user, or from a community of users.

If one of the predicted queries is what the searcher intended as the desired query, the searcher may select that predicted query and proceed without having to finish entering the desired query.

Alternatively, if the predicted queries do not reflect what the searcher had in mind, then the searcher can continue entering the desired search query, which could trigger one or more other sets of search suggestions.

Ranking User Submitted Previous Queries as Search Suggestions

The patent tells us that a few different processes may be used in ranking and ordering predicted search queries:

  • Predicted search queries may be ordered in accordance with a frequency of submission by a community of users
  • Time constraints may also be used with search queries ordered in accordance with the last time/date value that the query was submitted
  • Personalization information or community information may be used such as information about subjects, concepts or categories of information that are of interest to the user (from prior search or browsing information)
  • Personalization may also be from a group that the searcher is associated with or belongs to (a member or an employee.)
  • According to a first ranking criteria, such as predefined popularity criteria, and then possibly reordered if any of the predicted search queries match the user personalization information of the user, to place the matching predicted search queries at or closer to the top of the ordered set of predicted search queries
  • Information provided by the tracking component and the result selection log(s) might be used for ranking and ordering the predicted search queries. (click data, language-specific, and country-specific data.)
  • Processed click data (e.g., aggregated click data for a given query) could be used for ranking and ordering predicted search queries – or each query a score may be calculated by summing click data (e.g., weighted clicks, etc.) on documents associated with the query, and predicted queries may be ordered based upon the score (e.g., higher values representing better)

An Information Model Based On Previously Submitted Query Data to Obtain Search Suggestions Predictions

This model can be used to predict what query data might satisfy a searcher the most by looking at long click information. A timer can be used to track how long a user views or “dwells” on a document.

The amount of time is referred to as “click data”.

A longer time spent dwelling on a document, would be termed a “long click”, and can indicate that a user found the document to be relevant for their query.

A brief period viewing a document would be termed a “short click”, and can be interpreted as a lack of document relevance.

Click data is a count of each click type (e.g., long, medium, short) for a particular query and document combination.

This aggregated click data from model queries for a given document can be used to create a quality of result statistic for that document to enhance a ranking of that document.

Quality of result statistic can be a weighted average of the count of long clicks for a given document and query.

This description from the patent tells us about how click data might be stored in tuples:

A search engine (e.g., the search engine) or other processes may create a record in the model for documents that are selected by users in response to a query or a partial query. Each record within the model (herein referred to as a tuple: ) is at least a combination of a query submitted by users, a document reference selected by users in response to that query, and an aggregation of click data for all users that select the document reference in response to the query. The aggregate click data can be viewed as an indication of document relevance. In various implementations, model data can be location-specific (e.g. country, state, etc) or language-specific. For example, a country-specific tuple would include the country from where the user query originated from in whereas a language-specific tuple would include the language of the user query. Other extensions of model data are possible.

The model may also include Post-click behavior that has been tracked by the tracking component.

This patent does include a lot of information about how Google might use click tracking data when ranking search suggestion predictions. It tells us about the data that could be collected about clicks:

The information gathered for each click can include:

(1) the query (Q) the user entered,
(2) the document result (D) the user clicked on,
(3) the time (T) on the document,
(4) the interface language (L) (which can be given by the user),
(5) the country (C) of the user (which can be identified by the host that they use, such as www-store-co-uk to indicate the United Kingdom), and
(6) additional aspects of the user and session.

The time (T) can be measured as the time between the initial click through to the document result until the time the user comes back to the main page and clicks on another document result. Moreover, an assessment can be made about the time (T) regarding whether this time indicates a longer view of the document result or a shorter view of the document result, since longer views are generally indicative of quality for the clicked through result. This assessment about the time (T) can further be made in conjunction with various weighting techniques.

Beyond Long Clicks

We are also told that document views from the selections can be weighted based on viewing length information to produce weighted views of the document result.

So, rather than simply distinguishing long clicks from short clicks, a wider range of click through viewing times can be included in the assessment of result quality, where longer viewing times in the range are given more weight than shorter viewing times.

Predicted Search Suggestions

Google will sometimes display search suggestions using autocomplete and also based upon user data from previous queries from a searcher’s previous search history, or the history of someone whom the searcher may be associated with, such as a fellow member of an organization or a co-worker.

While results related to those previous queries were ranked based upon such things as relevance and backlinks, the search suggestions may include results that searchers spent long clicks upon, including long times viewing.

So pursuant to this patent, predictions about search suggestions chosen using autocomplete may best meet a searcher’s informational needs by being searches that include results remembered as resulting in long clicks and long viewing times.


Copyright © 2020 SEO by the Sea ⚓. This Feed is for personal non-commercial use only. If you are not reading this material in your news aggregator, the site you are looking at may be guilty of copyright infringement. Please contact SEO by the Sea, so we can take appropriate action immediately.
Plugin by Taragana

Search Suggestions from Previously Submitted Searcher Queries was originally posted by Video And Blog Marketing

Converting HTML to WordPress: Step by Step Tutorial

Web Development

Creating a website is no longer that difficult, but that wasn’t always the case.

In days gone by, if you weren’t a coding master you couldn’t even begin to think about creating a website for your business.

Thankfully, we live in the software age.

Thanks to the invention of website templates, software apps and automatic content management systems, website design has become a lot easier. There are dozens of tools that can do most of the heavy lifting when it comes to the design and coding of your website.

The main one, of course, being WordPress.

But what if you had your site designed years ago?

You’d still have to update the site via HTML coding. And that can be time-consuming at best and ridiculously expensive at worst.

So does that mean you’ll have to scrap your whole website and start from scratch?

Not at all.

It’s possible to convert your old HTML site code into a WordPress website.

Static HTML sites still exist, and they still have a place on the modern Web. But if you’re not a coding expert and you want to take charge of your site personally, it might be a good idea to move from static HTML to WordPress.

What’s more, this doesn’t have to be an overly complicated process. Methods vary from advanced coding to plugins.

We’ll explore those methods in this article.

So, how can you convert your static HTML website into WordPress? Read on to find out.

Why Would You Convert HTML Into a WordPress Theme?

If you already have a static HTML site, why would you want to convert that into a WordPress theme, to begin with?

Well, for starters, it’s far easier to manage.

If you’re not someone fluent in website coding, you’ll need developers to make any changes to your site.

Also, if you want to stay on top of SEO to be found in popular search engines, you’re going to need to update your site consistently.

That’s where WordPress comes into play.

WordPress is a content management system (CMS) designed so that anyone can use it, regardless of skill level. You won’t need to hire a developer for WordPress development, and changes are super simple.

You can often make changes with just a few clicks of your mouse (even over mobile).

But what does it mean when we say that you should convert your existing website into WordPress?

Converting your website into WordPress means taking your existing data from your current, static HTML site and transferring it into a WordPress theme.

There are three main ways to do this, which we will explain below.

1. Manual Conversion of HTML to WordPress

Because it is the most technical option on our list, manual conversion is not recommended for everyone.

Manual conversion uses your current site’s HTML code as a starting point. If you’re going to attempt this conversion method, it is recommended that you have some coding experience. Specifically, you should know HTML and CSS, as well as PHP.

Luckily, most of this process involves copy and paste, but it’s still complicated.

Here is a step-by-step guide to manual HTML/WordPress conversion:

Step 1: Create a New Theme Folder

The first thing that you’ll need to do is create a new theme folder on your desktop. Think of it as a directory folder on your computer. It serves the same purpose.

Now, go to the code editor and create text files. There are five different files you’ll want to create:

  1. Style.css
  2. Index.php
  3. Header.php
  4. Sidebar.php
  5. Footer.php

Step 2: Copy CSS Code

Next, you’ll have to copy the CSS coding from your old website onto a WordPress Style Sheet.

To do that, you’ll have to prepare the WordPress style sheet, which is the style.css file you created in the last step.

Copy and paste the CSS code from the old site’s source into that style sheet.

Then it’s time to fill out the various parts of the style sheet header for your new WordPress theme.

They are:

  • Theme Name – This can be anything you want.
  • Theme URL – The homepage information or site address.
  • Author – Your name.
  • Author URL – Link to the homepage you’re building.
  • Description – This part is an optional write-up on the theme that shows within the WordPress backend.
  • Version – Start with 1.0.
  • License, License URL, Tags – This part is only necessary if you’re going to submit the theme into the WordPress directory for others to use. If you’re keeping it for yourself, then don’t worry about it.

Here’s what that style sheet might look like:

Once you’re done with the header, paste the CSS code from the static HTML site into your file. Save the file in your theme folder and close it.

Step 3: Separate Existing HTML

WordPress uses PHP to access database information. As a result, your existing HTML code has to be chopped into separate pieces so that the WordPress CMS can properly string them together.

To do this, you’ll have to copy parts of the original HTML document into several different PHP files.

First, open your index.html file.

Go through the WordPress files that were created and copy that code into the following areas:

  1. Header.php – This entails everything from the beginning of your HTML code up to the main content area. Right before the section marked </head> you’ll have to copy and paste <?php wp_head();?>
  2. Sidebar.php – This is where you put all the code from the section marked <aside>
  3. Footer.php – This section starts at the end of the sidebar and goes up to the end of the file. Add a call for <?php wp_footer();?> before closing off the bracket with </body>.

Once you’ve done that, close the index.html file and save your other data to the theme folder.

Close all of the files except for header.php and index.php.

Step 4: Change the Header.php and Index.php Files for WordPress

Next, you’ll be changing the header.php and index.php to fit into WordPress’s format.

To do this, look for a link in the <head> section that looks like this:

<link rel=”stylesheet” href=”style.css”>.

Replace that link with this:

<link rel=”stylesheet” href=”<?php echo get_template_directory_uri(); ?>/style.css” type=”text/css” media=”all” />.

Now, save and close the header.php file. You’re done with it for the moment.

Open your index.php file. It should be empty.

Enter the following, precisely like this:

<?php get_header(); ?>

<?php get_sidebar(); ?>

<?php get_footer(); ?>

The space between the first and second lines of code is essential. This is where you will paste your Loop code. It’s a form of PHP used by WordPress for displaying posts:

After that, save and close the file. The basic theme should be ready and can now be added to a WordPress site.

Step 5: Screenshot and Upload

The last thing you’ll need to do is create a screenshot of your theme and upload it.

The screenshot will show a preview of your site in the WordPress backend.

Take this screenshot and crop it to 880×660 pixels. Save the file as a screenshot.png.

Now, add the screenshot to your theme folder.

It’s time to upload the theme to WordPress. Take the following five steps:

  1. Create a zip file.
  2. Go to WordPress.
  3. Select Appearance, Themes, and click Add New at the top.
  4. Click Upload Theme.
  5. Upload your zip file and click Install Now.

Once that’s done, you can activate the theme!

2. Converting HTML Through a WordPress Child Theme

Using a WordPress child theme to turn your original HTML into a CMS format gives you a lot more freedom and doesn’t require nearly the amount of technical know-how as the previous method.

It’s also the easiest and cheapest option for converting HTML to WordPress.

To do this, you’re going to use a ready-made theme as a jumping-on point instead of modeling it off your existing site.

It is possible to adjust the design of your WordPress parent theme so that it resembles the old site as much as possible.

That means you’ll be able to use WordPress while retaining the look and feel of the original site. There is no need to add WordPress features after because you’re building the new website on an existing theme.

Child themes are built on top of another theme, which is called the parent. The child theme modifies the parent theme in a way that fits your specific site.

Here is a step-by-step guide to converting your static HTML site into WordPress using a child theme.

Step 1: Choose a Theme

Before you can get started, you need to pick a theme. Try to find one that you like, but also resembles your existing design.

Install the theme on your WordPress site like you would any other theme. Just don’t activate it yet.

Step 2: Create a New Theme Folder

You’re going to create a new folder on your desktop, much like you did in the previous method.

Name the folder the same as the parent theme and add “-child” to the end of it. Remember, there should be no spaces in the name.

Step 3: Create a Style Sheet

This step is identical to the style sheet creation we went through in the previous method.

However, this time, you’re going to add a tag titled “template.” Make sure that you include the name of your parent theme. That is needed for the child theme to work.

Step 4: Create a Functions.php

Next, you’ll create a functions.php and inherit the parent styles for the child theme.

To do this, create a new file and call it functions.php. Make sure you start it off with <?php.

Now, input the following code:

function child_theme_enqueue_styles() {

$parent_style = ‘parent-style’;

wp_enqueue_style( $parent_style, get_template_directory_uri() . ‘/style.css’ );

wp_enqueue_style( ‘child-style’,

get_stylesheet_directory_uri() . ‘/style.css’,

array( $parent_style ),

wp_get_theme()->get(‘Version’)

);

}

add_action( ‘wp_enqueue_scripts’, ‘child_theme_enqueue_styles’ );

This code lets WordPress know to go to the parent theme and use the styles that are listed there for the child theme.

Step 5: Activate the Child Theme

You can now activate the child theme.

Before that, though, make sure that you take a screenshot to be featured on the WordPress backend.

Create a zip file with everything and add it all into WordPress, as we did in the previous method.

You’ll then be able to change the design to match your original HTML.

3. Import Your Content From HTML into WordPress Using a Plugin

This tactic is only recommended if you’re open to changing your site’s design. If you want an all-new look, using WordPress plugins can be a much easier road to travel.

Here is a step-by-step guide on how you can import your content from HTML into WordPress using plugins.

Step 1: Set Up a New Site

Start up your new site and install the WordPress theme of your choice. Make sure it’s a template that you like and is easily edited. You will need to change the appearance to match your branding.

Step 2: Install the Plugin

Now, it’s time to install the plugin that makes this possible. You’re going to search for a WordPress Plugin called HTML Import 2 and install it on your site.

Click on Install Now and then activate it.

Step 3: Upload Pages

Once the plugin is up and running, upload your pages to the same server as your WordPress installation.

Under the Files tab, you’ll enter the following information:

  • Directory to Import – This is the pathway you copied your existing HTML code to
  • Old site URL – The old URL is mostly there for redirect purposes. Enter the old URL of the site.
  • Default File – Enter your index.html.
  • File extensions to include – Put in the extensions of the files that will be imported.
  • Directories to exclude – Exclude anything from the old site that you don’t want to be carried over.
  • Preserve file names – The plugin will automatically use your file names as the new URL.

Once that’s done, go under the content tab and configure the HTML tag that holds your site’s content.

There are several tabs that you’ll have to familiarize yourself with:

  • Under the Title and Metadata tab, you’ll let the plugin know how your titles are marked in the HTML template.
  • The Custom Fields tab is where you put data that needs to be imported into custom fields.
  • On the Categories and Tags tab, you’ll assign categories to your imported content.
  • The tools screen is where you can go over some of the built-in tools found in the extension.

Once you’ve gone through every tab, save your settings, and click Import Files.

In Conclusion

If you have a static HTML site, it’s a good idea to switch over to a more effective content management system with proven functionality, like the WordPress platform.

Thanks to WordPress templates and the easy-to-use WordPress Dashboard, HTML to WordPress conversion will make your website easier to manage and a whole lot cheaper to maintain.

Converting HTML to WordPress: Step by Step Tutorial was originally posted by Video And Blog Marketing

Google Officially Releases May 2020 Core Update

Last updated on

Google Officially Releases May 2020 Core Update_Cover PhotoIt’s the time of the year again where Google releases its core update. However, times are different now because of the ongoing pandemic and we won’t know if this core update will negatively affect us. Being on top of everything right now is important since a lot of businesses are opting to go digital because of the ongoing pandemic. These give us SEOs and digital marketers a massive opportunity to bring quality services to more people who need it. But with core updates, success in the search industry is uncertain since unexpected variables and factors are constantly added in the SEO game. 

Google announced the May 2020 Core update through their Google SearchLiaison Twitter account a few hours ago. A follow-up tweet was made an hour after the original announcement that informed everyone that the May 2020 Core Update is now live and is rolling out steadily although it could take up to two weeks before it’s fully rolled out. 

Google May 2020 Core Update Announcement

What’s included in the May 2020 Core Update?

Google did not specify which changes the May 2020 core update and this is understandable since core updates do not focus on individual factors, rather they focus on the algorithm as a whole. Google once again referenced their blog post about what we should know about their core updates.

There are three possibilities that could happen once the core update has fully rolled out: 

  • It positively affects your site rankings and traffic
  • It negatively affects your site rankings and traffic
  • It doesn’t affect your site rankings and traffic

We’ll just have to wait and see how the core update affects you. But if it does negatively affect you, I’ve written a post on how to recover from being hit by the core update

The last core update was released in January 2020. This might seem like just a short time difference between the 2 updates, but I believe the May 2020 Core Update might have something to do with the massive changes in search behavior due to the pandemic. 

SERP Volatility

Lastly, I checked tools that included SERP volatility just for the sake of ensuring that there’s nothing major happening right now. Here’s what they looked like:

CognitiveSEO Signals

CognitiveSEO google ranking volatility screenshot

Around the end of April and the start of May, there was a spike with SERP volatility, but it went down during the past few days.

 

Accuranker’s Google Grump Rating

Accuranker Google Grump Rating screenshot

Compared with cognitiveSEO’s graph, Accuranker’s Google grump rating shows that the spike happened today. This could possibly be the result of the core update, but we still need a few days and more data to finalize this assumption.

 

SEMrush Sensor

SEMrush sensor screenshot

Unlike the other two tools, SEMrush sensor isn’t showing signs of massive spikes and volatility. 

 Key Takeaway

In conclusion, SERP volatility tools are not showing conclusive signs of the core update affecting the SERPs as of this moment. However, there is a high chance for this to change in the one or two weeks that is needed to roll out the full core update.

We just have to be ready to adapt to the situation and I believe if we know for ourselves that we’re not doing anything that’s worth penalizing, we shouldn’t worry about the occasional core updates that Google releases. What do you think about this core update? Comment it down below and let’s talk.

Google Officially Releases May 2020 Core Update was originally posted by Video And Blog Marketing

Distilled and Brainlabs Have Combined Forces: What’s Next For Us?

In case you missed it, back in February, Distilled merged into Brainlabs. This was the first step towards a joint vision of creating the agency model of the future: no silos, no departments, just brilliant teams centered around clients, allowing us to deliver best-in-class digital marketing. Since Distilled was acquired, Hanapin Marketing & Hero Conf have also merged with Brainlabs.

Now we’re a few months in, it’s time to reflect on everything that’s happened and take a quick look at where we are going. 

What does this mean for SEO consulting?

For our team, it is full steam ahead. We’re are still going to have offices in London, New York and Seattle, and with the addition of Hanapin in the US, we are going to be working to expand our SEO capabilities to even more offices in the US. We’re already running workshops for all the Brainlabs team and increasing their SEO knowledge and our SEO capacity.

We continue to offer expert consulting to our clients, but we now have access to more resources and data than ever before. And this goes both ways, we’ve already seen existing SEO clients taking advantage of Brainlabs’ PPC power and vice versa. 

There is a whole host of opportunities for all our clients to now run the SEO and Paid Media (Search and Social) capabilities, all under one roof. If you want to speak to our team you can still contact us via distilled.net/contact.

What does this mean for blog content?

We’ve been writing here for years, and yes, we actually mean 11 years (here’s one of our early efforts looking into generic vs local TLDs)!

We’ve always pushed our team to share their knowledge, findings and resources through our blog. We will continue to be publishing content here frequently for the immediate future. Over time we’ll be transitioning to the Brainlabs site (Will has already posted some thoughts), but we’ll make sure you know when we do.

In the meantime, make sure that you follow us and Brainlabs in all the right places. During the merger, our @distilled accounts have now become @SearchLove. Check us out on Twitter (SearchLove and Brainlabs) and  LinkedIn (SearchLove and Brainlabs).

What does this mean for SearchLove conferences?

In the midst of COVID-19, conferences have ground to a halt. Unfortunately, we’ve already had to postpone SearchLove San Diego, New York and London. Once we reach the new normal SearchLove will be back, bigger and better than ever before.

Our team is going to be headed up by Lynsey, who has been at the front of our conferences for many years. We’ll also be sharing our knowledge with the Hero Conf team and grabbing ideas from them to make our conferences better than ever!

We’ve already got new dates for SearchLove San Diego and are working on plans for our New York and London conferences. We’ll also be heading back to all three cities in 2021!

What does this mean for SearchPilot?

SearchPilot is our SEO A/B testing tool, formerly known as DistilledODN. It was spun out into an independent company before Distilled and SearchLove merged with Brainlabs. 

Our teams will continue to work closely together, as well as sharing test information and results. We’ll keep sharing these findings with you through the blog. 

You can keep following the SearchPilot story at searchpilot.com and by following them on Twitter and LinkedIn.

To wrap up

  1. Follow these accounts:

  1. If you haven’t already done so, sign up to our newsletter, and when the time comes we’ll take you with us on our Brainlabs journey.

Distilled and Brainlabs Have Combined Forces: What’s Next For Us? was originally posted by Video And Blog Marketing

How the Google Call Ads Update Can Maximize SEM

Last updated on

How the Google Call Ads Update Can Maximize SEM_Cover Photo

Starting out to advertise on Google is not easy. When you have finally decided on the course that you are going to take with your Search Engine Marketing (SEM), you have to take it upon yourself to maximize the space that Google has given you. This is also a cost-effective way to advertise because you can make sure that your budget will not burn without cause.

Recently, Google has announced that they have added website links to Call Ads. This ensures that your customers have options to engage with your business aside from picking up the phone to call. What does this mean for your SEM efforts? Find out here!

Website Links to Call Ads Update

Since the introduction of the Call-Only-Ads way back in 2015, this has given marketers the chance to segment the engagement of customers per channel that they are tracking. The call ads only extension gives your contact details in a straightforward manner for customers who would want to talk to your company instantly. This direct approach has done wonders in terms of engaging with customers since they would know how to get in touch with you immediately and in turn, your business can keep track of the conversions that your SEM effort can deliver in real-time.

Here’s the catch, since many customers can see and instantly engage with the contact detail, this proves can risk a high volume of calls. Failing to attend to the queue of calls can mean that customers will not repeat the engagement either because they have waited too long or they cannot reach you at all. Now, with the website links to call ads update coming into play, this gives them the chance to interact with your website instead of calling you up.

Google-Call-Ads>

<p< span=””>>Here is the mock-up of the Ad from Google. As you can see it looks like the basic Call Ad format, but with the option to visit the website below. Do take note that this is optional, not mandatory to have. In terms of user behavior, you can predict that they would go into the website and find other ways to address their inquiry. With this, you would not have any bad customer experience under your belt because of calls left unaddressed.

 

In terms of design, the ad is just a simple one but projecting the results that you can gain from this can give you a new perspective about the way that you strategically create ads for your business.

</p<>

Business implications for call ad extensions<!–h2>

The visibility that you can gain when you start marketing on the search engine is unmatched. With the right strategy and managed expectations for the ad, you can garner over a million impressions. If you are wondering how this can affect your ad performance and user behavior other than greater visibility, then here are some of the reasons why your business can use this type of extension to your advantage.

Immediate Engagement

The leads you get from your Google Ad can easily convert into a buyer since the visibility of this ad is capable of appearing on most smartphone devices. With this, the leads that are ready to buy will go straight to you. Having this type of ad extension is a no-nonsense approach to marketing since it is a direct approach for customers to get in touch with you.

More Quality Leads</h2>

Quality leads are those that engage with your business with a clear intention to buy. Seeing you as the authority of that industry, enough to make them call you, is a huge plus in getting quality leads. The call ad extension can help you see that the customer is willing and able to engage with your business because you already lock in their intention to buy the second they call you. For this section, Google says that “If your business relies on phone calls for new sales, you’ll get fewer accidental calls and more qualified leads.” Having a call ad will help you feel secure in the customer’s intention to buy since they would not dial your number without being interested in what you have to offer.

Increased information channels for customers

The new update validates customer intention to buy because according to Google, “you can now engage consumers who may only be interested in visiting your site, and may not have previously interacted with your brand.” On top of that, you would not fall victim to a bad customer experience because this will lessen the risk of having long hold times. Before talking with them personally, the “visit to website” option can help them browse around and get to know your business better.

Key Takeaway

For businesses that advertise on Google, the end goal is to have leads and convert them into customers. Online visibility is gaining importance now more than ever so it is vital that you maximize your efforts to secure this for your business. With the right tools and strategies, leads and conversions will come naturally to you. The best way to maximize these efforts is to keep track of these updates and see how it would work for your online presence. What are the Google Ads updates that worked for your marketing efforts? Comment down below.

How the Google Call Ads Update Can Maximize SEM was originally posted by Video And Blog Marketing