Google’s Rich Results Test Tool is Now Out of Beta

Last updated on

In the Official Google Webmaster Central blog, Google announced that the Rich Results Test tool is now officially out of beta and is now supporting all of Google’s rich results feature.

Google also announced that the Structured Data Testing tool is going to be deprecated. It is still available for a short time but there is no specific date yet on when Google will officially remove it.

Moving forward, Google urges webmasters to use the Rich Results Test tool to validate all structured data markups. Here’s what you need to know about the Rich Results Test tool.

How does the Rich Results Test Work?

Rich results (formerly known as rich card and rich snippets) are special types of search results that are different from the regular blue links in Google. It uses structured data to identify which type of rich results a page is eligible for.

In the Rich Results Test tool, there are two options to validate; either via the specific URL of the page you want to check or a code snippet. There’s also an option to select a user-agent the tool will use but right now, only Google Smartphone bot is available.

Once the tool is done analyzing the page, it will show the types of rich results a page or code is valid for. It will also identify any errors that may hinder your webpage from appearing in the rich results. You could also click on “Preview Results” and it will show how your page will look as a Rich Result.

Here are the types of rich results that are available in the Google search results right now:

  • Article
  • Book
  • Breadcrumb
  • Carousel
  • Course
  • Critic Review
  • Dataset
  • Employer Aggregate Rating
  • Event
  • Fact Check
  • FAQ
  • How-to
  • Image License Metadata
  • Job Posting
  • Job Training
  • Local Business
  • Logo
  • Movie
  • Estimated Salary
  • Podcast
  • Product
  • Q&A
  • Recipe
  • Review Snippet
  • Sitelink Searchbox
  • Software App
  • Speakable
  • Subscription and Paywalled Content
  • Video

If you want to learn more about how each type of rich result looks and the proper structured data markup to be eligible for them, you could check out Google’s Search Gallery.

Reactions from SEOs

In the official Google Webmasters Twitter account announcement, many SEOs were disappointed about the tool and the planned depreciation of the structured data testing tool.

As pointed out by Barry Adams and Digitaleer, the Structured Data Testing tool can validate any type of structured data markup, not only those that are valid for rich results. In my opinion, the structured data testing tool was a great tool on its own because I use markups that are not included yet in the rich results list.

The Structured Data Testing tool was also great for debugging. SEOs have noticed that compared to it, the Rich Results Test tool would usually vague error messages.

It seems like the release did not go smoothly as Google planned but I do believe that they will add more rich results types and make it available in the tool. And with the deprecation of the Structured Data Testing Tool, I do hope that they move the feature of being able to validate all types of structured data markup in the Rich Results Test.

Key Takeaway

Since rich results are visually appealing to users, they tend to get a large chunk of traffic from the search results similar to featured snippets. That is why it is important that you check the structured data markups of your website and make sure they are eligible for rich results.

Prior to the release from beta, the rich results test tool can only validate a handful of structured data markup but it is great news that it can now validate all rich results types. I highly recommend that you make use of this tool and grab the opportunity to rank for rich results.

 

Google’s Rich Results Test Tool is Now Out of Beta was originally posted by Video And Blog Marketing

Website Rankings Drop: Identifying the Reasons Why

Last updated on

Website Rankings Drop Identifying the Reason Why_Cover Photo

It’s a common occurrence for SEOs and webmasters that track their rankings to experience sudden website ranking drops. Since Google’s algorithm is continuously updating, SERP volatility is common and affects hundreds or even thousands of websites in different industries. The problem with undiagnosed ranking drops is the decrease in your website’s search visibility. This could lead to less traffic, and for business websites, this leads to fewer conversions and profit. Identifying the cause of the rankings drop is tantamount to fixing it and, hopefully, recovering your rankings. So, how do you identify the cause of your website’s rankings drop?

Knowing Where to Look

The first step to identifying the cause of your rankings drop is to identify the variables that changed. This is because a drop in rankings won’t happen for no reason. There will always be a reason for changes in rankings (positive and negative) and this usually happens due to a change in a factor/s that affects your page’s rankings. 

This may sound easy but it’s not. There are numerous factors that affect a page’s rankings, some we can identify and some are still a mystery. That is why determining the exact factor why your rankings dropped is easier said than done. 

But here’s what you should do: Limit the scope of your search. Since you’ll just tire yourself out investigating factors that no one knows about, limit your scope to factors that YOU have control over. So, once you have limited your scope of search, this is when you’ll investigate deeper. 

The way we do it in SEO Hacker is to identify changes in 3 major places that contain a variety of factors that can affect our rankings. Then, we’ll dive deeper and deeper until we find the changes that COULD contribute to the website rankings drop and we either change them back or improve them further than they initially were. So, where exactly are these 3 major places?

INSIDE Your Website

Since we need to limit the scope of our search to things we have control over, it makes sense that we start with the page/website immediately. You have to first identify if the website is experiencing ranking drops in multiple keywords or if you only experience a drop for a specific keyword that a single page is ranking for. Take a look at this example:

Page ranking drop screenshot

The rankings drop happened to a single page that has been consistently ranking positively for numerous months. We automatically limited the scope of our search to the page itself. After determining that technical factors such as title tag, h1, etc. have been changed, we took it a step further and changed it for the better (not revert it back to the original). After a few days waiting for it to be recrawled, It ranked higher than it has ranked for the past few months. 

Since we limited our scope, easily found out the changes made and fixed them immediately, we didn’t waste our energy and time checking other factors that would have turned out to be unrelated to the drop. So, we highly suggest that you start INSIDE your website.

But what happens if nothing changed inside the page or website? Then, you check out other factors outside your website. 

OUTSIDE Your Website

The scope for this is limited to your search competitors overtaking you on the results page. This usually happens when they update their page to serve better intent, contain more information and become more relevant and useful for the users, or an overall upgrade of their website’s technical factors. Here’s an example:

page ranking drop outside the website screenshot

This happened recently and we’re confident that we did not change anything on the page that’s ranking for this specific keyword. So, we started our diagnosis in the search results of the page and that’s where we saw it. Competitors updated their pages with new content, a new design, or a fully revamped one. All of these changes enabled them to overtake our page in the search results which resulted in us having a lower ranking than before. 

So, it’s important for you as an SEO or webmaster to take note of the competitors you have for a specific keyword. Taking note of their content, their page design, and every other factor that they could change will enable you to know which of them they changed. This allows you to quickly adapt and regain your rankings as soon as possible.

Google Algorithm Update

We continuously report confirmed Google algorithm updates and its details to help our readers be able to understand why they experienced a change in rankings immediately. It is well-known in the industry that an algorithm update causes ranking volatility in the search results. A massive drop in rankings usually means that you’re doing something wrong and Google is penalizing you for that. 

Sticking to proven white-hat strategies and having an in-depth understanding of Google’s search algorithm will enable you to avoid penalties every time Google updates their algorithms since what you’re doing is in-line with what they want you to do. 

Key Takeaway

Identifying the reason behind your website or page’s rankings drop is important to achieving search success. Knowing where to pour your efforts in and properly invest the time and energy to investigate them enables you to grow as an SEO and webmaster.

Problem-solving is a necessity in the SEO industry since it continuously changes and gives us no choice but to adapt our strategies to reach success, that’s hopefully, for the long-term. How do you investigate whenever your website’s ranking drops? Let me know in the comments below!

Website Rankings Drop: Identifying the Reasons Why was originally posted by Video And Blog Marketing

How Google Might Rank Image Search Results

Changes to How Google Might Rank Image Search Results

We are seeing more references to machine learning in how Google is ranking pages and other documents in search results.

That seems to be a direction that will leave what we know as traditional, or old school signals that are referred to as ranking signals behind.

It’s still worth considering some of those older ranking signals because they may play a role in how things are ranked.

As I was going through a new patent application from Google on ranking image search results, I decided that it was worth including what I used to look at when trying to rank images.

Images can rank highly in image search, and they can also help pages that they appear upon rank higher in organic web results, because they can help make a page more relevant for the query terms that page may be optimized for.

Here are signals that I would include when I rank image search results:

  • Use meaningful images that reflect what the page those images appear on is about – make them relevant to that query
  • Use a file name for your image that is relevant to what the image is about (I like to separate words in file names for images with hyphens, too)
  • Use alt text for your alt attribute that describes the image well, and uses text that is relevant to the query terms that the page is optimized for) and avoid keyword stuffing
  • Use a caption that is helpful to viewers and relevant to what the page it is about, and the query term that the page is optimized for
  • Use a title and associated text on the page the image appears upon that is relevant for what the page is about, and what the image shows
  • Use a decent sized image at a decent resolution that isn’t mistaken for a thumbnail

Those are signals that I would consider when I rank image search results and include images on a page to help that page rank as well.

A patent application that was published this week tells us about how machine learning might be used in ranking image search results. It doesn’t itemize features that might help an image in those rankings, such as alt text, captions, or file names, but it does refer to “features” that likely include those as well as other signals. It makes sense to start looking at these patents that cover machine learning approaches to ranking because they may end up becoming more common.

Machine Learning Models to Rank Image Search Results

Giving Google a chance to try out different approaches, we are told that the machine learning model can use many different types of machine learning models.

The machine learning model can be a:

  • Deep machine learning model (e.g., a neural network that includes multiple layers of non-linear operations.)
  • Different type of machine learning model (e.g., a generalized linear model, a random forest, a decision tree model, and so on.)

We are told more about this machine learning model. It is “used to accurately generate relevance scores for image-landing page pairs in the index database.”

We are told about an image search system, which includes a training engine.

The training engine trains the machine learning model on training data generated using image-landing page pairs that are already associated with ground truth or known values of the relevance score.

The patent shows an example of the machine learning model generating a relevance score for a particular image search result from an image, landing page, and query features. In this image, a searcher submits an image search query. The system generates image query features based on the user-submitted image search query.

Rank Image Search Results includes Image Query Features

That system also learns about landing page features for the landing page that has been identified by the particular image search result as well as image features for the image identified by that image search result.

The image search system would then provide the query features, the landing page features, and the image features as input to the machine learning model.

Google may rank image search results based on various factors

Those may be separate signals from:

  1. Features of the image
  2. Features of the landing page
  3. A combining the separate signals following a fixed weighting scheme that is the same for each received search query

This patent describes how it would rank image search results in this manner:

  1. Obtaining many candidate image search results for the image search query
  2. Each candidate image search result identifies a respective image and a respective landing page for the respective image
  3. For each of the candidate image search results processing
  • Features of the image search query
  • Features of the respective image identified by the candidate image search result
  • Features of the respective landing page identified by the candidate image search result using an image search result ranking machine learning model that has been trained to generate a relevance score that measures a relevance of the candidate image search result to the image search query
  • Ranking the candidate image search results based on the relevance scores generated by the image search result ranking machine learning model
  • – Generating an image search results presentation that displays the candidate image search results ordered according to the ranking
    – Providing the image search results for presentation by a user device

    Advantages to Using a Machine Learning Model to Rank Image Search Results

    If Google can rank image search query pairs based on relevance scores using a machine learning model, it can improve the relevance of the image search results in response to the image search query.

    This differs from conventional methods to rank resources because the machine learning model receives a single input that includes features of the image search query, landing page, and the image identified by a given image search result to predicts the relevance of the image search result to the received query.

    This process allows the machine learning model to be more dynamic and give more weight to landing page features or image features in a query-specific manner, improving the quality of the image search results that are returned to the user.

    By using a machine learning model, the image search engine does not apply the same fixed weighting scheme for landing page features and image features for each received query. Instead, it combines the landing page and image features in a query-dependent manner.

    The patent also tells us that a trained machine learning model can easily and optimally adjust weights assigned to various features based on changes to the initial signal distribution or additional features.

    In a conventional image search, we are told that significant engineering effort is required to adjust the weights of a traditional manually tuned model based on changes to the initial signal distribution.

    But under this patented process, adjusting the weights of a trained machine learning model based on changes to the signal distribution is significantly easier, thus improving the ease of maintenance of the image search engine.

    Also, if a new feature is added, the manually tuned functions adjust the function on the new feature independently on an objective (i.e., loss function, while holding existing feature functions constant.)

    But, a trained machine learning model can automatically adjust feature weights if a new feature is added.

    Instead, the machine learning model can include the new feature and rebalance all its existing weights appropriately to optimize for the final objective.

    Thus, the accuracy, efficiency, and maintenance of the image search engine can be improved.

    The Rank Image Search results patent application can be found at

    Ranking Image Search Results Using Machine Learning Models
    US Patent Application Number 16263398
    File Date: 31.01.2019
    Publication Number US20200201915
    Publication Date June 25, 2020
    Applicants Google LLC
    Inventors Manas Ashok Pathak, Sundeep Tirumalareddy, Wenyuan Yin, Suddha Kalyan Basu, Shubhang Verma, Sushrut Karanjkar, and Thomas Richard Strohmann

    Abstract

    Methods, systems, and apparatus including computer programs encoded on a computer storage medium, for ranking image search results using machine learning models. In one aspect, a method includes receiving an image search query from a user device; obtaining a plurality of candidate image search results; for each of the candidate image search results: processing (i) features of the image search query and (ii) features of the respective image identified by the candidate image search result using an image search result ranking machine learning model to generate a relevance score that measures a relevance of the candidate image search result to the image search query; ranking the candidate image search results based on the relevance scores; generating an image search results presentation; and providing the image search results for presentation by a user device.

    The Indexing Engine

    The search engine may include an indexing engine and a ranking engine.

    The indexing engine indexes image-landing page pairs, and adds the indexed image-landing page pairs to an index database.

    That is, the index database includes data identifying images and, for each image, a corresponding landing page.

    The index database also associates the image-landing page pairs with:

    • Features of the image search query
    • Features of the images, i.e., features that characterize the images
    • Features of the landing pages, i.e., features that characterize the landing page

    Optionally, the index database also associates the indexed image-landing page pairs in the collections of image-landing pairs with values of image search engine ranking signals for the indexed image-landing page pairs.

    Each image search engine ranking signal is used by the ranking engine in ranking the image-landing page pair in response to a received search query.

    The ranking engine generates respective ranking scores for image-landing page pairs indexed in the index database based on the values of image search engine ranking signals for the image-landing page pair, e.g., signals accessed from the index database or computed at query time, and ranks the image-landing page pair based on the respective ranking scores. The ranking score for a given image-landing page pair reflects the relevance of the image-landing page pair to the received search query, the quality of the given image-landing page pair, or both.

    The image search engine can use a machine learning model to rank image-landing page pairs in response to received search queries.

    The machine learning model is a machine learning model that is configured to receive an input that includes

    (i) features of the image search query
    (ii) features of an image and
    (iii) features of the landing page of the image and generate a relevance score that measures the relevance of the candidate image search result to the image search query.

    Once the machine learning model generates the relevance score for the image-landing page pair, the ranking engine can then use the relevance score to generate ranking scores for the image-landing page pair in response to the received search query.

    The Ranking Engine behind the Process to Rank Image Search Results

    In some implementations, the ranking engine generates an initial ranking score for each of multiple image—landing page pairs using the signals in the index database.

    The ranking engine can then select a certain number of the highest-scoring image—landing pair pairs for processing by the machine learning model.

    The ranking engine can then rank candidate image—landing page pairs based on relevance scores from the machine learning model or use those relevance scores as additional signals to adjust the initial ranking scores for the candidate image—landing page pairs.

    The machine learning model would receive a single input that includes features of the image search query, the landing page, and the image to predict the relevance (i.e., relevance score, of the particular image search result to the user image query.)

    We are told that this allows the machine learning model to give more weight to landing page features, image features, or image search query features in a query-specific manner, which can improve the quality of the image search results returned to the user.

    Features That May Be Used from Images and Landing Pages to Rank Image Search Results

    The first step is to receive the image search query.

    Once that happens, the image search system may identify initial image-landing page pairs that satisfy the image search query.

    It would do that from pairs that are indexed in a search engine index database from signals measuring the quality of the pairs, and the relevance of the pairs to the search query, or both.

    For those pairs, the search system identifies:

    • Features of the image search query
    • Features of the image
    • Features of the landing page

    Features Extracted From the Image

    These features can include vectors that represent the content of the image.

    Vectors to represent the image may be derived by processing the image through an embedding neural network.

    Or those vectors may be generated through other image processing techniques for feature extraction. Examples of feature extraction techniques can include edge, corner, ridge, and blob detection. Feature vectors can include vectors generated using shape extraction techniques (e.g., thresholding, template matching, and so on.) Instead of or in addition to the feature vectors, when the machine learning model is a neural network the features can include the pixel data of the image.

    Features Extracted From the Landing Page

    These aren’t the kinds of features that I usually think about when optimizing images historically. These features can include:

    • The date the page was first crawled or updated
    • Data characterizing the author of the landing page
    • The language of the landing page
    • Features of the domain that the landing page belong to
    • Keywords representing the content of the landing page
    • Features of the links to the image and landing page such as the anchor text or source page for the links
    • Features that describe the context of the image in the landing page
    • So on

    Features Extracted From The Landing Page That Describes The Context of the Image in the Landing Page

    The patent interestingly separated these features out:

    • Data characterizing the location of the image within the landing page
    • Prominence of the image on the landing page
    • Textual descriptions of the image on the landing page
    • Etc.

    More Details on the Context of the Image on the Landing Page

    The patent points out some alternative ways that the location of the image within the Landing Page might be found:

    • Using pixel-based geometric location in horizontal and vertical dimensions
    • User-device based length (e.g., in inches) in horizontal and vertical dimensions
    • An HTML/XML DOM-based XPATH-like identifier
    • A CSS-based selector
    • Etc.

    The prominence of the image on the landing page can be measured using the relative size of the image as displayed on a generic device and a specific user device.

    The textual descriptions of the image on the landing page can include alt-text labels for the image, text surrounding the image, and so on.

    Features Extracted from the Image Search Query

    The features from the image search query can include::

    • Language of the search query
    • Some or all of the terms in the search query
    • Time that the search query was submitted
    • Location from which the search query was submitted
    • Data characterizing the user device from which the query was received
    • So on

    How the Features from the Query, the Image, and the Landing Page Work Together

    • The features may be represented categorically or discretely
    • Additional relevant features can be created through pre-existing features (Relationships may be created between one or more features through a combination of addition, multiplication, or other mathematical operations.)
    • For each image-landing page pair, the system processes the features using an image search result ranking machine learning model to generate a relevance score output
    • The relevance score measures a relevance of the candidate image search result to the image search query (i.e., the relevance score of the candidate image search result measures a likelihood of a user submitting the search query would click on or otherwise interact with the search result. A higher relevance score indicates the user submitting the search query would find the candidate image search more relevant and click on it)
    • The relevance score of the candidate image search result can be a prediction of a score generated by a human rater to measure the quality of the result for the image search query

    Adjusting Initial Ranking Scores

    The system may adjust initial ranking scores for the image search results based on the relevance scores to:

    • Promote search results having higher relevance scores
    • Demote search results having lower relevance scores
    • Or both

    Training a Ranking Machine Learning Model to Rank Image Search Results

    The system receives a set of training image search queries
    For each training image search query, training image search results for the query that are each associated with a ground truth relevance score.

    A ground truth relevance score is the relevance score that should be generated for the image search result by the machine learning model (i.e., when the relevance scores measure a likelihood that a user would select a search result in response to a given search query, each ground truth relevance score can identify whether a user submitting the given search query selected the image search result or a proportion of times that users submitting the given search query select the image search result.)

    The patent provides another example of how ground-truth relevance scores might be generated:

    When the relevance scores generated by the model are a prediction of a score assigned to an image search result by a human, the ground truth relevance scores are actual scores assigned to the search results by human raters.

    For each of the training image search queries, the system may generate features for each associated image-landing page pair.

    For each of those pairs, the system may identify:

    (i) features of the image search query
    (ii) features of the image and
    (iii) features of the landing page.

    We are told that extracting, generating, and selecting features may take place before training or using the machine learning model. Examples of features are the ones I listed above related to the images, landing pages, and queries.

    The ranking engine trains the machine learning model by processing for each image search query

    • Features of the image search query
    • Features of the respective image identified by the candidate image search result
    • Features of the respective landing page identified by the candidate image search result and the respective ground truth relevance that measures a relevance of the candidate image search result to the image search query

    The patent provides some specific implementation processes that might differ based upon the machine learning system used.

    Take Aways to Rank Image Search Results

    I’ve provided some information about what kinds of features Google May have used in the past in ranking Image search results.

    Under a machine learning approach, Google may be paying more attention to features from an image query, features from Images, and features from the landing page those images are found upon. The patent lists many of those features, and if you spend time comparing the older features with the ones under the machine learning model approach, you can see there is overlap, but the machine learning approach covers considerably more options.


    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

    How Google Might Rank Image Search Results was originally posted by Video And Blog Marketing

    How Google Might Rank Image Search Results

    Changes to How Google Might Rank Image Search Results

    We are seeing more references to machine learning in how Google is ranking pages and other documents in search results.

    That seems to be a direction that will leave what we know as traditional, or old school signals that are referred to as ranking signals behind.

    It’s still worth considering some of those older ranking signals because they may play a role in how things are ranked.

    As I was going through a new patent application from Google on ranking image search results, I decided that it was worth including what I used to look at when trying to rank images.

    Images can rank highly in image search, and they can also help pages that they appear upon rank higher in organic web results, because they can help make a page more relevant for the query terms that page may be optimized for.

    Here are signals that I would include when I rank image search results:

    • Use meaningful images that reflect what the page those images appear on is about – make them relevant to that query
    • Use a file name for your image that is relevant to what the image is about (I like to separate words in file names for images with hyphens, too)
    • Use alt text for your alt attribute that describes the image well, and uses text that is relevant to the query terms that the page is optimized for) and avoid keyword stuffing
    • Use a caption that is helpful to viewers and relevant to what the page it is about, and the query term that the page is optimized for
    • Use a title and associated text on the page the image appears upon that is relevant for what the page is about, and what the image shows
    • Use a decent sized image at a decent resolution that isn’t mistaken for a thumbnail

    Those are signals that I would consider when I rank image search results and include images on a page to help that page rank as well.

    A patent application that was published this week tells us about how machine learning might be used in ranking image search results. It doesn’t itemize features that might help an image in those rankings, such as alt text, captions, or file names, but it does refer to “features” that likely include those as well as other signals. It makes sense to start looking at these patents that cover machine learning approaches to ranking because they may end up becoming more common.

    Machine Learning Models to Rank Image Search Results

    Giving Google a chance to try out different approaches, we are told that the machine learning model can use many different types of machine learning models.

    The machine learning model can be a:

    • Deep machine learning model (e.g., a neural network that includes multiple layers of non-linear operations.)
    • Different type of machine learning model (e.g., a generalized linear model, a random forest, a decision tree model, and so on.)

    We are told more about this machine learning model. It is “used to accurately generate relevance scores for image-landing page pairs in the index database.”

    We are told about an image search system, which includes a training engine.

    The training engine trains the machine learning model on training data generated using image-landing page pairs that are already associated with ground truth or known values of the relevance score.

    The patent shows an example of the machine learning model generating a relevance score for a particular image search result from an image, landing page, and query features. In this image, a searcher submits an image search query. The system generates image query features based on the user-submitted image search query.

    Rank Image Search Results includes Image Query Features

    That system also learns about landing page features for the landing page that has been identified by the particular image search result as well as image features for the image identified by that image search result.

    The image search system would then provide the query features, the landing page features, and the image features as input to the machine learning model.

    Google may rank image search results based on various factors

    Those may be separate signals from:

    1. Features of the image
    2. Features of the landing page
    3. A combining the separate signals following a fixed weighting scheme that is the same for each received search query

    This patent describes how it would rank image search results in this manner:

    1. Obtaining many candidate image search results for the image search query
    2. Each candidate image search result identifies a respective image and a respective landing page for the respective image
    3. For each of the candidate image search results processing
    • Features of the image search query
    • Features of the respective image identified by the candidate image search result
  • Features of the respective landing page identified by the candidate image search result using an image search result ranking machine learning model that has been trained to generate a relevance score that measures a relevance of the candidate image search result to the image search query
  • Ranking the candidate image search results based on the relevance scores generated by the image search result ranking machine learning model
  • – Generating an image search results presentation that displays the candidate image search results ordered according to the ranking
    – Providing the image search results for presentation by a user device

    Advantages to Using a Machine Learning Model to Rank Image Search Results

    If Google can rank image search query pairs based on relevance scores using a machine learning model, it can improve the relevance of the image search results in response to the image search query.

    This differs from conventional methods to rank resources because the machine learning model receives a single input that includes features of the image search query, landing page, and the image identified by a given image search result to predicts the relevance of the image search result to the received query.

    This process allows the machine learning model to be more dynamic and give more weight to landing page features or image features in a query-specific manner, improving the quality of the image search results that are returned to the user.

    By using a machine learning model, the image search engine does not apply the same fixed weighting scheme for landing page features and image features for each received query. Instead, it combines the landing page and image features in a query-dependent manner.

    The patent also tells us that a trained machine learning model can easily and optimally adjust weights assigned to various features based on changes to the initial signal distribution or additional features.

    In a conventional image search, we are told that significant engineering effort is required to adjust the weights of a traditional manually tuned model based on changes to the initial signal distribution.

    But under this patented process, adjusting the weights of a trained machine learning model based on changes to the signal distribution is significantly easier, thus improving the ease of maintenance of the image search engine.

    Also, if a new feature is added, the manually tuned functions adjust the function on the new feature independently on an objective (i.e., loss function, while holding existing feature functions constant.)

    But, a trained machine learning model can automatically adjust feature weights if a new feature is added.

    Instead, the machine learning model can include the new feature and rebalance all its existing weights appropriately to optimize for the final objective.

    Thus, the accuracy, efficiency, and maintenance of the image search engine can be improved.

    The Rank Image Search results patent application can be found at

    Ranking Image Search Results Using Machine Learning Models
    US Patent Application Number 16263398
    File Date: 31.01.2019
    Publication Number US20200201915
    Publication Date June 25, 2020
    Applicants Google LLC
    Inventors Manas Ashok Pathak, Sundeep Tirumalareddy, Wenyuan Yin, Suddha Kalyan Basu, Shubhang Verma, Sushrut Karanjkar, and Thomas Richard Strohmann

    Abstract

    Methods, systems, and apparatus including computer programs encoded on a computer storage medium, for ranking image search results using machine learning models. In one aspect, a method includes receiving an image search query from a user device; obtaining a plurality of candidate image search results; for each of the candidate image search results: processing (i) features of the image search query and (ii) features of the respective image identified by the candidate image search result using an image search result ranking machine learning model to generate a relevance score that measures a relevance of the candidate image search result to the image search query; ranking the candidate image search results based on the relevance scores; generating an image search results presentation; and providing the image search results for presentation by a user device.

    The Indexing Engine

    The search engine may include an indexing engine and a ranking engine.

    The indexing engine indexes image-landing page pairs, and adds the indexed image-landing page pairs to an index database.

    That is, the index database includes data identifying images and, for each image, a corresponding landing page.

    The index database also associates the image-landing page pairs with:

    • Features of the image search query
    • Features of the images, i.e., features that characterize the images
    • Features of the landing pages, i.e., features that characterize the landing page

    Optionally, the index database also associates the indexed image-landing page pairs in the collections of image-landing pairs with values of image search engine ranking signals for the indexed image-landing page pairs.

    Each image search engine ranking signal is used by the ranking engine in ranking the image-landing page pair in response to a received search query.

    The ranking engine generates respective ranking scores for image-landing page pairs indexed in the index database based on the values of image search engine ranking signals for the image-landing page pair, e.g., signals accessed from the index database or computed at query time, and ranks the image-landing page pair based on the respective ranking scores. The ranking score for a given image-landing page pair reflects the relevance of the image-landing page pair to the received search query, the quality of the given image-landing page pair, or both.

    The image search engine can use a machine learning model to rank image-landing page pairs in response to received search queries.

    The machine learning model is a machine learning model that is configured to receive an input that includes

    (i) features of the image search query
    (ii) features of an image and
    (iii) features of the landing page of the image and generate a relevance score that measures the relevance of the candidate image search result to the image search query.

    Once the machine learning model generates the relevance score for the image-landing page pair, the ranking engine can then use the relevance score to generate ranking scores for the image-landing page pair in response to the received search query.

    The Ranking Engine behind the Process to Rank Image Search Results

    In some implementations, the ranking engine generates an initial ranking score for each of multiple image—landing page pairs using the signals in the index database.

    The ranking engine can then select a certain number of the highest-scoring image—landing pair pairs for processing by the machine learning model.

    The ranking engine can then rank candidate image—landing page pairs based on relevance scores from the machine learning model or use those relevance scores as additional signals to adjust the initial ranking scores for the candidate image—landing page pairs.

    The machine learning model would receive a single input that includes features of the image search query, the landing page, and the image to predict the relevance (i.e., relevance score, of the particular image search result to the user image query.)

    We are told that this allows the machine learning model to give more weight to landing page features, image features, or image search query features in a query-specific manner, which can improve the quality of the image search results returned to the user.

    Features That May Be Used from Images and Landing Pages to Rank Image Search Results

    The first step is to receive the image search query.

    Once that happens, the image search system may identify initial image-landing page pairs that satisfy the image search query.

    It would do that from pairs that are indexed in a search engine index database from signals measuring the quality of the pairs, and the relevance of the pairs to the search query, or both.

    For those pairs, the search system identifies:

    • Features of the image search query
    • Features of the image
    • Features of the landing page

    Features Extracted From the Image

    These features can include vectors that represent the content of the image.

    Vectors to represent the image may be derived by processing the image through an embedding neural network.

    Or those vectors may be generated through other image processing techniques for feature extraction. Examples of feature extraction techniques can include edge, corner, ridge, and blob detection. Feature vectors can include vectors generated using shape extraction techniques (e.g., thresholding, template matching, and so on.) Instead of or in addition to the feature vectors, when the machine learning model is a neural network the features can include the pixel data of the image.

    Features Extracted From the Landing Page

    These aren’t the kinds of features that I usually think about when optimizing images historically. These features can include:

    • The date the page was first crawled or updated
    • Data characterizing the author of the landing page
    • The language of the landing page
    • Features of the domain that the landing page belong to
    • Keywords representing the content of the landing page
    • Features of the links to the image and landing page such as the anchor text or source page for the links
    • Features that describe the context of the image in the landing page
    • So on

    Features Extracted From The Landing Page That Describes The Context of the Image in the Landing Page

    The patent interestingly separated these features out:

    • Data characterizing the location of the image within the landing page
    • Prominence of the image on the landing page
    • Textual descriptions of the image on the landing page
    • Etc.

    More Details on the Context of the Image on the Landing Page

    The patent points out some alternative ways that the location of the image within the Landing Page might be found:

    • Using pixel-based geometric location in horizontal and vertical dimensions
    • User-device based length (e.g., in inches) in horizontal and vertical dimensions
    • An HTML/XML DOM-based XPATH-like identifier
    • A CSS-based selector
    • Etc.

    The prominence of the image on the landing page can be measured using the relative size of the image as displayed on a generic device and a specific user device.

    The textual descriptions of the image on the landing page can include alt-text labels for the image, text surrounding the image, and so on.

    Features Extracted from the Image Search Query

    The features from the image search query can include::

    • Language of the search query
    • Some or all of the terms in the search query
    • Time that the search query was submitted
    • Location from which the search query was submitted
    • Data characterizing the user device from which the query was received
    • So on

    How the Features from the Query, the Image, and the Landing Page Work Together

    • The features may be represented categorically or discretely
    • Additional relevant features can be created through pre-existing features (Relationships may be created between one or more features through a combination of addition, multiplication, or other mathematical operations.)
    • For each image-landing page pair, the system processes the features using an image search result ranking machine learning model to generate a relevance score output
    • The relevance score measures a relevance of the candidate image search result to the image search query (i.e., the relevance score of the candidate image search result measures a likelihood of a user submitting the search query would click on or otherwise interact with the search result. A higher relevance score indicates the user submitting the search query would find the candidate image search more relevant and click on it)
    • The relevance score of the candidate image search result can be a prediction of a score generated by a human rater to measure the quality of the result for the image search query

    Adjusting Initial Ranking Scores

    The system may adjust initial ranking scores for the image search results based on the relevance scores to:

    • Promote search results having higher relevance scores
    • Demote search results having lower relevance scores
    • Or both

    Training a Ranking Machine Learning Model to Rank Image Search Results

    The system receives a set of training image search queries
    For each training image search query, training image search results for the query that are each associated with a ground truth relevance score.

    A ground truth relevance score is the relevance score that should be generated for the image search result by the machine learning model (i.e., when the relevance scores measure a likelihood that a user would select a search result in response to a given search query, each ground truth relevance score can identify whether a user submitting the given search query selected the image search result or a proportion of times that users submitting the given search query select the image search result.)

    The patent provides another example of how ground-truth relevance scores might be generated:

    When the relevance scores generated by the model are a prediction of a score assigned to an image search result by a human, the ground truth relevance scores are actual scores assigned to the search results by human raters.

    For each of the training image search queries, the system may generate features for each associated image-landing page pair.

    For each of those pairs, the system may identify:

    (i) features of the image search query
    (ii) features of the image and
    (iii) features of the landing page.

    We are told that extracting, generating, and selecting features may take place before training or using the machine learning model. Examples of features are the ones I listed above related to the images, landing pages, and queries.

    The ranking engine trains the machine learning model by processing for each image search query

    • Features of the image search query
    • Features of the respective image identified by the candidate image search result
    • Features of the respective landing page identified by the candidate image search result and the respective ground truth relevance that measures a relevance of the candidate image search result to the image search query

    The patent provides some specific implementation processes that might differ based upon the machine learning system used.

    Take Aways to Rank Image Search Results

    I’ve provided some information about what kinds of features Google May have used in the past in ranking Image search results.

    Under a machine learning approach, Google may be paying more attention to features from an image query, features from Images, and features from the landing page those images are found upon. The patent lists many of those features, and if you spend time comparing the older features with the ones under the machine learning model approach, you can see there is overlap, but the machine learning approach covers considerably more options.


    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

    How Google Might Rank Image Search Results was originally posted by Video And Blog Marketing

    CRO Audits and How You Can Use it For E-Commerce

    Last updated on

    How CRO Can Target E-Commerce Demand 2020_Cover Photo

    Conversion Rate Optimization is one of the areas that businesses should look into if they want to continue empowering growth in their respective industries. A winning strategy starts with a well-formed audit from the CRO specialist that targets contingencies and issues that can come up in the pursuit of astounding conversions. An online presence will only be treated as a nice-to-have for your business if it does not serve any revenue-generating benefit.

    With this, we have to look at the overall climate in business today. While it is unfortunate that some of the brands we have come to know and love have decided to close their doors because of the drastic effect that the pandemic enforced on their business, there is still hope in coping up with the new normal.

    With most people staying home, retail stores can be seen at their twilight and e-commerce continues to drive the consumer industries as online payments have been proven to be the safest alternative to normal shopping trips. If you are planning on upgrading your site to reap the benefits that e-commerce can give for your business, it is imperative that you know the flow of CRO and its impressive effect on revenue.

    Starting a Solid CRO Strategy

    A CRO audit would require you to know your business to a tee. You need to be able to illustrate what your business can offer and come up with materials that will allow users to see that you are easily accessible to provide something of value to them. Usually, this can come in the form of one or more of the following:

    • Catalogs
    • Brochures
    • E-Books
    • Brand Bible
    • Other Downloadable Materials for Products/Services

    Depending on the overall structure of your site and your direction with e-commerce, you can explore different options on offering these materials. You can create a pop-up that will allow a user to connect with you through promotions or other offers. The key to this is to make it worthy of attention but non-imposing so you have to insert an option for the user to opt-out of viewing the pop-up.

    pop up toyota

    Take a closer look at your buttons as well. This a subtle but essential part of User Experience (UX) that not a lot of CRO specialists take advantage of.

    sticky header toyota

    Aside from a pop-up, you can also create a sticky header on your site that offers information or a promotion that a user would surely not miss. At the end of the day, research will do you good. Research more about what your business can offer, what potential leads can use to close a deal with you, and know more about design principles that will make your CRO stand out.

    CRO and SEO: Your Tools for E-Commerce Success

    Conversion Rate Optimization and Search Engine Optimization go hand in hand. They may be alike in terms of the goal of making a positive impact on a website, but they serve separate purposes for it as well. With CRO, you should also look at your efforts the same way you scrutinize a site for SEO. Look at your site’s assets and its weak points as well. Illustrate its missing elements and plan how you can design your strategy from there.

    With tools at your disposal, you can get to the root of your problem with leads and conversions right away. Qeryz is one of the most reliable tools that I use for CRO. With this, I can see what customers are saying about a business and what they expect out of it. You can also target semantic elements here since it can help you see the queries that they are looking for in relation to your site. With a customer survey tool, you can hear it straight from the customer what exactly they are looking for which saves you a lot of time in doing guessing games for market research.

    If you are interested to explore Qeryz, you can look at its benefits here.

    Another CRO tool that I can recommend is Inspectlet. For heatmaps or other movement indicators on the site, you can rely on this tool to detect user behavior as accurately as possible. With this, you can see session recordings of users; what pages they visited, those that increased their on-page time. You can see in real-time what products or services the users are looking at which can give way to more opportunities for your business to leverage the visibility that these products are gaining.

    Having tools to help you with your CRO is a great help in determining what areas of UI and UX need more work. On top of the previously mentioned benefit, you can also see how user behavior plays in maintaining an e-commerce site. Coming up with a strategy for CRO will be an advantage for long-term returns.

    Key Takeaway

    The key to coming up with a powerful CRO audit is to know your site well. Study what its strengths are and what it can bring to the table. Your online visibility will be put to good use if you explore CRO as a viable strategy for digital marketing. In this age of resilience, this is an option that you shouldn’t dare miss. What are the tools you use for your own Conversion Rate Optimization practice? Comment them down below.

    CRO Audits and How You Can Use it For E-Commerce was originally posted by Video And Blog Marketing

    How to Create a Social Media Content Calendar

    Social Media

    If you’re a social media manager, you need to plan out your entire month of posts. Never leave it to the last minute, trying to come up with new and exciting post ideas day by day.

    By creating a social media content calendar, you ensure that all of your posts adhere to your content strategy.

    But creating a social media content calendar takes skill and know-how. You need to understand what a social media calendar is, why it’s useful, and how you can go about making one.

    What is a Social Media Content Calendar?

    If you’re a social media content manager, the first thing that you need to know about your content calendar is that it is an incredibly helpful tool that will let you plan out your social postings on a monthly basis.

    Here’s an example of a social media content calendar:

    If you’re not already using a social media management tool, then a social media content calendar is usually an excel file, with various tabs assigned to different social platforms. Remember, the posting requirements differ based on the platform you’re using. Facebook has different restrictions than Twitter. Twitter restrictions will be different from Instagram, and so on.

    One of the best features of a social media content calendar is that it can help you keep track of themed days of the month.

    That can mean two different things.

    You can create your own themed days, like Inspirational Mondays or Workspace Wednesdays. It’s a great idea to have these days because you can create a hashtag around them and see if they catch fire.

    The other meaning behind themed days of the month is national days. Every day of the year is “National (Something) Day.” There’s everything from National Bunny Day to National Bread Day. Creating social media content based around these holidays can generate interest. Keeping track of them in your content calendar is a great way to plan around them.

    (Image Source)

    The social media content calendar helps you plan out your content and schedule ahead of time using a system. It can also help you keep track of the images you’re going to share, organizing them week by week and day by day.

    Download Our Social Media Calendar Template

    Enter your email and instantly get a weekly social media content calendar template.

    If you are human, leave this field blank.
    Download Calendar Template

    Why Should You Use a Social Media Content Calendar?

    Every business that wants to maximize their social media presence should be using a social media content calendar. There’s a reason that 92% of content marketing professionals use social calendars.

    A social media content calendar can help you address the challenge of declining organic social media reach in two primary ways.

    If you’ve tried to create a social media presence for your business, there’s one thing you probably noticed straight away…

    Social media sites are crowded.

    Every business is trying to get a piece of the social media pie. That’s because social media is the ideal marketing hunting ground. Practically every demographic has some kind of social media presence, from teens to senior citizens and everyone in between.

    Because of this, the social media platforms themselves have seen dollar signs where business marketing is concerned. Facebook founder Mark Zuckerberg doesn’t make any money if your organic post goes viral. Facebook and other social media giants want you to pay to boost your posts and purchase ads. That’s how they make their money.

    (Image Source)

    This was confirmed by Zuckerberg himself in 2018, when he publicly stated that Facebook users would be seeing fewer organic posts from businesses and brands.

    One of the only ways to get noticed with organic posting is to approach your social media strategy with a plan.

    And a content calendar helps you keep track of that plan.

    The social media calendar will help you keep to schedule, a crucial element if you’re looking to make a splash in the social media world. Remember, the more consistently you post, the better your exposure will be. Customers get used to regular postings, and will seek you out in time.

    But in order to establish that routine, you have to post on a regular basis.

    And you need to plan out your content platform by platform.

    You need to keep track of what you are posting and where. The calendar represents an easy record that you can go back and look through.

    This record also helps you figure out how you’re doing in terms of social content. If you have a huge influx of comments and follows, you’ll want to remember what you did in order to duplicate your results.

    The calendar can help you determine where these spikes occurred and what content was going on each platform.

    It’s also important to keep track of the time that you’re posting, and check your results. Remember, different platforms experience high traffic at different times. That means all of your posts should not go out at the same time across every channel.

    Finally, another great benefit of a social media content calendar is that it can be easily shared with team members for convenient collaboration.

    Download Our Social Media Calendar Template

    Enter your email and instantly get a weekly social media content calendar template.

    If you are human, leave this field blank.
    Download Calendar Template

    Creating a Social Media Content Calendar

    Creating a social media content calendar takes a lot of effort. There are a number of steps you’ll have to take before you write the first line of content. There’s a massive amount of work that goes into creating the perfect social media strategy. Your calendar is just one part of that overall project.

    Let’s look at the steps needed to make a social media content calendar one at a time and see where to begin.

    Step 1: Information Gathering

    You could consider the information-gathering stage of your social media content strategy to be an audit of your current social media needs.

    In order to improve, you first have to figure out where you are, where you’ve been, and what’s preventing you from getting to where you want to be.

    The first step to a good social media audit is determining which platforms you’re currently using and what level of success you’re seeing for all of them. You should take the time to review your results for at least the last six months and use them to rank all of your social media platforms in terms of success.

    When you see all of this laid out before you, there’s a decision to be made. Moving forward, you need to decide if you’re going to eliminate anything from your lineup. For example, if you’re seeing no movement from Google+ whatsoever, try to figure out why that is. Are you not posting enough? Are you posting the wrong content? Are you posting at the wrong times?

    Or is your audience not on that platform in large numbers?

    (Image Source)

    If you find that no matter what you do, Google+ just isn’t going to deliver, then it might be time to shut down your Google+ account, or at least put less effort into it.

    It’s important to feed the strong, so if you’re getting a lot of interaction on Instagram, that’s where more of your effort (and by extension, marketing budget) should be going.

    When you look at all of your social media accounts side by side, you should be contrasting and comparing them to one another.

    Are these accounts uniform in terms of branding?

    You need to present a united front when it comes to marketing. If your Twitter account looks wildly different from your Facebook, particularly your profile and header images, that could be jarring for potential customers. You want all of your branding to blend. That includes your social media sites and your website.

    Do you have access to every account?

    Sometimes when an account is neglected for a long period of time, an organization could misplace the login information. It’s important that you don’t have any duplicate pages sitting out there with outdated information. If customers were to search for your business and find that page, it makes you look bad.

    When auditing your social media, you also need to figure out what kind of audience you’re reaching (if any) and what kind of audience you want to reach.

    You should have a good handle on your ideal customer. Are you creating content specifically with them in mind?

    Information on your demographic should be included in the social media content calendar. You can set that up as a static reminder in the heading of your document. That way you never forget who you’re speaking to when creating content.

    Step 2: Demographic Study

    When creating social media content to go in your calendar, there are a few key questions that you need to ask.

    Who are your customers?

    What do they want?

    Once you understand the demographics you serve, you’ll be able to create better content that is geared specifically toward their interests. Once you’ve done that, it should be a simple matter to get them to participate in discussions.

    Does your existing demo differ platform by platform? You need to figure out where the eyes of your audience are. Once you know that, put more effort into placing your content there. Don’t expect your customers to come to you. You have to go to them.

    All of the content in your calendar should be tailored to each platform that you’re posting on. This is not a one size fits all approach. Content created for Twitter will differ from content created for Facebook.

    Once you know what your audience is looking for and where they are looking, you can create tabs in your calendar file for each specific social media site.

    Step 3: Create a regular schedule

    The purpose of your social media calendar is to keep your posting to a schedule. Before you can start filling out content, you first have to decide what that schedule is.

    How often are you going to post? That’s a question that can only be answered by understanding your audience. You don’t want to annoy them by popping up in their feed too often.

    What time will you post?

    As we mentioned above, this should differ on each platform, but (in general) here’s a good place to start:

    (Image Source)

    Step 4: Decide on Your Content Voice

    What kind of content do you want to share? Should it be serious? Silly? Both? If you’re going to create both, what is the ratio you’re going for?

    You also need to decide if you’re going to be creating posts that are designed more for engagement, and how often you are going to “shill” your products or services, if ever.

    If your demographic seems open to creating user-generated content, you should invite that. Set up themed days for them to participate in. Try something like “Furry Friend Fridays,” where your Fans or Followers can post pictures of their pets.

    Step 5: Create a Database of Content

    Collect a library of useful articles, images, and concepts to share with your audience. Keep them in a folder and be ready to pop them into the calendar at a moment’s notice.

    When you list this content, make sure that you mark any time-specific information. You don’t want to sit on a good article only to have so much time pass that it’s no longer relevant.

    Step 6: Add the Content

    Once you know when to post and what you’re going to post, it’s time to input all of that information into your calendar. You should have a set time where you do this every month.

    For example, you could take the last week of the month to create a calendar for the next month. This ensures that you won’t forget to create the content and have to scramble to be ready for the month ahead.

    Consistency is important in your preparation as well as your posting.

    Step 8: Share Your Calendar

    Make sure that you’re sharing your calendar with your supervisors and colleagues. That presents an opportunity to get feedback from the rest of your team.

    Make adjustments to your content strategy based on calendar feedback. Sometimes it’s hard to critique yourself, and the people you work with might see things from a different perspective. Ask the sales team for advice. They speak with your audience on a more frequent basis and should have a good handle on what they’re looking for.

    Helpful Tools

    If you need a helping hand in getting started with your social media calendar, there are a number of online tools that you can turn to that will help you get moving in the right direction.

    You’re going to want to use a pre-made template for your first calendar.

    Here is a list of four templates that could help you get started.

    In Conclusion

    All up and coming social media content managers need to develop a system.

    And the most effective and widely-used system in play is a social media content calendar.

    By inputting all of your social information ahead of time and keeping it organized, you’ll start seeing increased engagement and higher levels of brand awareness in no time.

    Download Our Social Media Calendar Template

    Enter your email and instantly get a weekly social media content calendar template.

    If you are human, leave this field blank.
    Download Calendar Template

    How to Create a Social Media Content Calendar was originally posted by Video And Blog Marketing

    How to Get in Google News

    Last updated on

    Google News is the go-to platform for millions of users worldwide for local and international news. It is the largest news aggregator in the world. It collects fresh news articles from around the web and serves it to users depending on relevancy, location, interests, and more. 

    Appearing in Google News provides a lot of benefits for website owners such as more website traffic and additional monetization options. It also opens up your website to a broader audience since it is available on various devices across different countries. And unlike Google Search, websites are served to a user through a feed rather than a search results page.

    Although Google collaborates with the news industry to make sure that only high-quality news appears on the platform, Google News is not limited to major news companies. As long as your website is a news website, whether it’s current events or industry news, you can appear in Google News and enjoy its benefits.

    Google News Eligibility

    I’ve received a lot of questions before from readers on how they get their website to appear in Google News. A few years ago, your website needs to be verified by Google News first before your content appeared on the platform. However, Google removed that requirement. Now, all you have to do is produce high-quality content if you’re a publisher. Sounds easy right?

    Well, to be more specific, the content that you have to publish needs to meet Google News content policies. Once you meet these criteria, your website is automatically considered to appear in the Top Stories section or News section.

    This method is quite difficult though because it would be hard to know if your website is really appearing in Google News.

    Another method is to submit your website to Google Publisher Center.

    Getting in Google News through Google Publisher Center

    Google Publisher Center is the combination of the old Google News Producer and Publisher Console. Google doesn’t require a website to be submitted in Google Publisher Center but it does have benefits such as control over your brand and content that appears in Google News.

    Do take note that submitting your content and RSS feeds on Google Publisher Center won’t mean that your content is guaranteed to appear in Google News. Also, keep in mind that the website you are going to submit to Google Publisher Center should be a verified website in your Google Search Console account.

    Step 1: Fill in the details needed

    Once you go to Google Publisher Center, simply click on Add Publication and enter the name of your website. The next screen will ask you for basic information about your websites such as description, website URL, and location. It is also important that you place the Google Analytics tracking ID in this section so you could track how much traffic you are getting from Google News.

    Step 2: Submit your content

    The Content section allows multiple ways to submit your content in Google News and customize your Google News feed by dividing them into multiple sections. You can either:

    • Submit RSS feed
    • Enter specific page URL
    • Submit YouTube video or playlist; or
    • Personalized feed from Google News

    Step 3: Upload images

    The Images section is where you upload your website’s square logo and wide logos. Make sure that the images you upload are high-quality and fit the specifications indicated.

    Step 4: Ads management (optional)

    This step is useful for websites that are running ads through Google AdSense. You could also control in this section how much ads Google sells on your website for users coming from Google News.

    Step 5: Submit for review

    The Review & Publish section will indicate if there is any missing information. Once you’re done filling in all the necessary details, you can now submit your website for review. 

    Keep in mind that approval may take 2 to 4 weeks according to Google. I think this depends on the number of websites they are reviewing. In my case, SEO Hacker got approved in 2 weeks and content that I published immediately started appearing in Google News. While you’re at it, why don’t you drop by SEO Hacker’s Google News page and follow us!

    How do I know How Much Traffic I got from Google News?

    In the Google Analytics account of your Google Publisher verified website, go to Acquisition, and check Referral traffic sources. If you are getting clicks from Google News, you should see news.google.com.

    How to Get More Traffic from Google News

    What most people don’t know is that Google News also uses an algorithm to rank websites in Google News. It uses the following factors:

    • Relevance of Content
    • Prominence
    • Authoritativeness
    • Freshness
    • Location
    • Language

    Users also have their own “For You” sections and Google News uses the users’ interests for rankings and usability.

    I also wrote an article before on How to Optimize for Google News and I highly recommend that you check that out as well so you can get more success in Google News.

    Key Takeaway

    Google News is a great initiative by Google for both users and web publishers. Achieving success may take time similar to growing traffic from organic search results but you should take any opportunities that you can get for you to be able to grow.

    In SEO, content is king. And it is more truthful in Google News. Always publish high-quality original content that is relevant to your audience and you will surely start getting higher placements in Google News.

    How to Get in Google News was originally posted by Video And Blog Marketing

    Understanding Semantic Search to Boost Your SEO

    Last updated on

    Semantic Search How to Improve Your SEO_Cover Photo

    Before 2013, SEO was virtually a straightforward endeavor. You had to incorporate the exact keywords inside your content and the important places such as title tag, H1, meta description, then build a high number of relevant quality links to those pages to ensure ranking on the first page. However, the same could not be said today. Search engines evolved in how they understand searches and queries to the point that the same strategies we used almost a decade ago are not enough now.

    Today, a deep understanding of the meaning behind the keywords, content that effectively answers the context of the keywords, and the intent behind it is what succeeds in the search results. All of these were brought about by the age of semantic search. Let’s dive deeper.

    What is Semantic Search?

    Semantic search, simple enough, is the search engine’s way to have a deeper understanding of a user’s queries by analyzing and connecting the meaning and relationship between the words and their context. This in turn enables search engines to find the best possible relevant pages to display even if the user’s query wasn’t straightforward. Here’s an example:

    quick brown fox google search screenshot

    This specific search is pertaining to the well-known phrase “The quick brown fox jumps over the lazy dog”. Google understands the context between the words used in the query and effectively answers it by knowing the relationship between the animal (quick brown fox) that jumped over the dog. Now, if semantic search was nonexistent, Google might have shown me results that showed random animals jumping over dogs. 

    This just goes to show how far search engines have come to understanding the queries made by users. However, much like other machines, search engines are also limited to their own knowledge. Here’s another example:

    chris evans google search screenshot

    Now, most of us know that the wizard in the Avengers is Doctor Strange, played by Benedict Cumberbatch. But when I searched for it, the answer Google gave me is Chris Evans, the actor that played Captain America. This isn’t the correct answer, far from it. But this is what happens when I change a word in my query:

    sorcerer in avengers google search screenshot

    The search result actually shows the correct answer – Benedict Cumberbatch being the actor that played Doctor Strange. The snippet isn’t as accurate as we’d like but it still does answer the question. So, if you focus on the query I made, I changed the word “wizard” to “sorcerer” and Google actually gave me a fairly accurate answer. Why did this happen? The reason behind this is easier to infer than you might think. 

    Through the use of Knowledge graphs, introduced back in 2012, Google has an extensive database of public information and the entities contained inside them such as the people involved, their birthdays, parents, etc. Now, through the knowledge graphs that Google has, they use what’s contained inside it to better understand queries. The inference we can make is that in Google’s database, Doctor Strange (and the actor that plays the role) is known as a sorcerer and not a wizard. This, possibly, could be the correct answer to why Google was able to give me a more accurate answer than the one I previously searched for. 

    To summarize, Google uses the following to better understand a user’s query:

    • Search Intent
    • Context
    • Relationship and connection between the words

    All of these things, coupled with your search history and behavior, affect how Google understands your queries and which results they display. Aside from Knowledge Graphs, Google released a series of updates over the years to make semantic search more accurate and effective.

    Algorithm Updates that Affect Semantic Search

    Hummingbird

    In 2013, Google released its Hummingbird update to help with understanding complex search queries. This update was released to reward the pages that match the query’s context rather than the page matching the words inside the query. 

    To simplify it more, Google rewarded pages that answered the user’s query rather than rewarding the pages that just contained repeated mentions of the words in the query.

    Rankbrain

    2 years after Hummingbird, Google released Rankbrain; a machine learning system that’s considered a ranking factor and analysis AI. Rankbrain shared the same purpose with Hummingbird but with the edge of its machine-learning integration.

    Through Rankbrain, Google was able to deliver more relevant results by interpreting the user’s through their location, behavior, and the words they used in the query. 

    BERT

    This is the newest addition to the updates Google Released. BERT is Google’s way of better understanding longer queries and more complex sentences by processing the words and their context & meaning in relation to the other words contained in the search query. 

    Through the BERT model, Google aims to deliver more accurate and relevant results to “longer, more conversational queries”. By the end of 2019, the BERT model affects 10% of all queries made in Google. This might not sound like a large number, but consider how many users Google has and how many times they search per day. 

    Understanding Semantic Search Leads to Better SEO

    Semantic search isn’t a small thing that we can disregard. It’s a massive process that changed an entire industry and it will only progress toward better things as the years go by. 

    SEO on the other hand is a practice that revolves around adapting all our strategies and tactics to hopefully reach the first page of an ever-growing search engine landscape. So, with semantic search, it’s only logical that our strategies change to better fit the current landscape. Some of the major changes we made in SEO Hacker include:

    • Changing our focus from keywords to topics

    It’s not about just writing about a specific keyword anymore. Much like the queries that users search for, keywords can mean a lot of things. That is why shifting our focus from a keyword-centric content strategy to a topically-relevant content framework did us wonders. 

    The primary difference between a keyword and a topic is their scope. A keyword is only limited to the meaning it holds while a topic will include multiple keywords under it. Knowing what the topic is and being able to write about it and its corresponding subtopics is an essential part of creating high-quality, valuable content today. 

    A great example is writing about title tags. If we were to solely focus on keywords, we would create pages that target “how to change a page’s title tag”, “best title tag length”, “how to optimize a title tag”, etc. 

    But if we focus on writing about title tag as a topic then we would write a page about “The Ultimate Guide to Title Tags” or “All You Need To Know About Title Tags” where the content talks about the subtopics such as “how to change a page’s title tag”, “best title tag length”, and “how to optimize a title tag”.

    Of course, you would need to know which subtopics are relevant to the users. You can’t write about everything about the topic at hand because that will not only waste your time and energy, but it will also discourage the user from reading everything as well. Using the example I used above, I wouldn’t include things such as “the history of title tag” or “who popularized the term title tag” simply because users wouldn’t care about those subtopics. 

    • Understanding and serving search intent

    Have you ever experienced ranking highly in the search results for a specific keyword, then you start noticing that you’re slowly ranking lower even though you know for yourself that the page ranking for the keyword is fully optimized and has valuable content? We’ve experienced that for a good number of our clients and the most common reason is that we’re not properly serving the intent of the keyword and the users. 

    Search intent is easily understood on paper but when applied in the search results, can be tricky and at times confusing. 

    The best way to mitigate this is to manually search for the keyword you’re ranking for and check the pages that are ranking higher than you. How does their content differ from your own? Is the context of their content different from your own? What kind of pages are ranking higher than you? Are they product category pages? Blog posts? Homepages? – these are questions you need to ask yourself when you’re slipping in rankings. 

    Simply put, your page should be something that Google would WANT to display on the first page of the search results for that specific keyword, and the best way to find it is to check what EXACTLY is Google displaying on the first page. 

    Key Takeaway

    Outdated SEO strategies and below-average content aren’t enough for the search engines AND users. Semantic Search has changed the SEO game for the better and it actually serves users more relevant and accurate answers than before. 

    Once you have a firm grasp of semantic search, then adapting to it will come easier than you think, especially if you’ve already made it a point to publish high-quality and relevant content. 

    Do you have any questions about semantic search? Comment it down below and let’s talk!

    Understanding Semantic Search to Boost Your SEO was originally posted by Video And Blog Marketing

    Google Ads Release Clickbait Policy

    Last updated on

    Google Releases New Policy on Clickbait Ads_Cover Photo

    No one is innocent from using clickbait copy, even top digital marketers sometimes use sensational anecdotes in their content. This is a problem because if you built your strategy on this lazy technique, then it will bite you in the ass before you know it. That time has come now. Google made an announcement last week in which they introduced a new Clickbait Ads policy which aptly addresses the Misrepresentation sections of their Ad policies.

    If you are planning on starting your ad with, “This client loved our business, what happened next will shock you,” or if you are planning on using this kind of copy in any of your Google Ads, content, don’t even think about it. It will be a waste of your money and worse, Google will not run your ads at all. Although this talks about Google Ads policies, I believe this should also be the same for Organic Search Optimization since creating too much clickbait content will only harm your rankings if an algorithm update explicitly tackles this area.

    Now, we are going to talk about why you should care about this new Ad policy, what can you do about it if you are guilty of using it for your ads, and what to do to avoid this.

    Implications of the New Policy

    Subheading Photo_Ban

    Together with the introduction of the new policy on clickbait ads, Google has also published a post on their Advertising Policies Help page on the Google Support website pertaining to account pausing action. They have notified advertisers about a new enforcement action for those who violate Google Ads policies. Before, if you violate a policy, the worst thing that can happen to you is your ads won’t run and that’s it. You’re still free to create a campaign for other Ad Groups, but Google noted this month that if you were investigated for a policy violation, your whole account will be temporarily paused and this also applies to advertisers who fail to verify their identity.

    As the new clickbait policy rolls out in July, think about the missed opportunities that you can incur if you violate this policy. You would be disabled from running ads and who knows what will happen if your offense is a grave enough to land you in Google’s crosshairs.

    Clickbait will not be an unbelievable copywriting strategy

    There is no good way to use clickbait. It deserves the bad reputation that it gets. Let me just put this out there, this is not a “copywriting hack” that seasoned marketers should teach the young blood of the industry. This just encourages lazy writing and an attitude of your work being just “okay”. Yes, it dominates most of the ads and even some organic search results but that doesn’t mean that if you can’t beat them, you join them.

    Each new piece of content that you are going to release for your audience deserves a well-thought-out execution. If you are going to write for people, make sure that you give something of value to them and this naturally falls into the goals that businesses have for Google Ads. You cannot expect people to become leads if they feel duped by clicking on your ad. You can claim that you can help them with their problem right away, they only need to click on the link, but the chances that they would be unsatisfied with your service or product is still there so why should you risk tarnishing your reputation just to exhaust your ads budget?

    Here are some key points you need to keep in mind to optimize your search ad for the clickbait policy:

    • For YMYL sites, avoid imposing your business to your audience’s lifestyle in a threatening manner with copies like, “You need this today or your business is going to fail.” This just sets you up for more trouble of violating the misrepresentation policy because the ‘Clickbait Ads’ policy also flags “ads that use negative life events such as death, accidents, illness, arrests or bankruptcy to induce fear, guilt or other strong negative emotions to pressure the viewer to take immediate action” Stop being edgy, just write a comprehensive copy that does not mislead your potential client.
    • This may sound extreme but I have seen some advertisers who write copies or launch display campaigns with “ads which use depictions of severe stress, pain, fear or shock to promote a product or service.” This is way before Google’s crackdown on these types of content but back then, it is an extreme way of pressuring someone to buy from a business – by inducing a negative personal connotation. Ease up on the aggressive approach, just be honest and highlight the strengths of your business in your advertisements.
    • “You won’t believe what happened next…” has been a part of the internet ecosphere for as long as I can remember. Think about the content that you are putting out there, does it properly represent your business or the growth that you are aiming for? Is the click more valuable than quality content or a conversion leading to a possible sale? You have to make the clicks matter.

    Optimizing Search Ads for Conversion Growth

    To end, I would like to say that clickbait will always be there but if you are smart enough as a marketer, you would not contribute to the misrepresentation that it can perceive itself to be. Always write with the reader in mind. Is it going to contribute to the value of their business, lifestyle, or interests? Is it geared towards professional growth? Surround yourself with these ideas and you will not dare write another clickbait Ad again.

    Google Ads Release Clickbait Policy was originally posted by Video And Blog Marketing

    Takeaways from Google’s Webspam Report 2019

    Last updated on

    Earlier this month, Google released its annual Webspam Report in the Official Google Webmaster Blog for 2019 where they discuss their fight against webspam. Webspam refers to webpages that try to manipulate Google’s algorithm or use techniques that are deceptive to users.

    As someone who is devoted to doing white-hat SEO, I personally look forward to this every year. This is not only a good way of getting in the know on what Google is doing to make their website a safer place for users but also it is a great way to avoid trouble for your website.

    Image from Google Webmaster Blog

    Google Webspam 2019 Highlights:

    • Google discovers 25 billion pages every day that is spammy and manipulative.
    • According to Google, 99% of web traffic from search results lead to spam-free experience.
    • Google was able to suppress link spam more. 90% of link spam techniques were caught by their systems.
    • Manipulative link building techniques such as paid links and link exchanges have been made less effective.
    • Hacked spam is still a challenge and Google is still finding more ways on how they can better alert website owners that their website has been hacked and help them recover from it.
    • Google is continuously improving its machine learning systems and combining it with its ‘manual enforcement capabilities’ to identify and prevent spam from getting to the search results.
    • There was an increasing amount of websites using auto-generated and scraped content but they were able to reduce the impact of these sites on users by 60% more from 2018.
    • Google received 230,000 manual reports of webspam from users and was able to resolve 83% of it.
    • Google sent out 90 million error messages to website owners via Google Search Console. 4.3 million of these messages are about manual actions.

    My Thoughts on this year’s report

    Google has gone a long way in improving its systems in detecting webspam. They are catching more and more websites and penalizing them but unfortunately, a lot of people are still doing it. Google has a lot on its plate. 25 billion spammy webpages every day is mind-blowing but it’s really great that they don’t get to the users anymore.

    It is also obvious that Google has doubled down on link spam and user-generated spam and it’s safe to say that it is related to the new link attributes. For us SEOs, it gives us more reasons to use the rel=”sponsored” and rel=”ugc” tags to avoid getting penalized.

    Website hacking is a persisting problem and it is definitely a huge one. It is good news that it is more stable the past year and Google is finding more ways to detect hacked sites and website owners. But I think it us up to the website owners to help solve this problem. Investing in your website’s security is crucial and you should not leave it to Google to protect your website.

    How can you help as an SEO?

    As SEOs, let us be more responsible and respond to Google’s call in helping them combat webspam. We have a better understanding of the world wide web and we can easily spot webspam. I believe that what Google is doing is for the better of the internet and I believe we play a huge role in the fight.

    230,000 webspam reports in the previous year is lower than I expected. But it may also be a good sign since Google is doing a great job in preventing webspam from appearing on search results.

    If you find pages in the search results that are spammy, deceptive, and manipulative, you can use the Report Webspam tool in Google Seach Console to report the webpage. There is also a Chrome Extension that you can download from the Chrome Web Store so you could easily report webspam. Make sure to use these tools responsibly and only when necessary. 

    Takeaways from Google’s Webspam Report 2019 was originally posted by Video And Blog Marketing