Google seems to be testing new font styles for the search results pages. Here is some screen shots of the current font that I see, followed by screen shots from some folks on Twitter noticing a new font style.

Here is what I see (click to enlarge):

Here is what @shemiadhikarath sees:

Here is what @jackalswitch79 sees:

Google is always testing, here is one more test that is ongoing.

Forum discussion at Twitter.

SEO is alive and well, but it has definitely changed over the years. If your search strategy hasn’t evolved to account for algorithm updates, then you’re probably not seeing the returns you want. After all, there isn’t a one-size-fits-all approach to SEO. The best strategies are continually reassessed and refined overtime to meet unique organizational objectives. Agility will serve you better than a comprehensive plan, especially right now amid the coronavirus pandemic. 

That’s not to say there aren’t guidelines for modern SEO. Your approach should include a steady stream of fresh, high-quality content (including well-written long-form content that adds real value to customers), a strong branding strategy, an exceptional user experience that highly meets the needs of visitors, and a strong backlink portfolio. Above all else, your strategy should prioritize your visitors’ experiences. 

Why, if you may ask?

Google ultimately ranks websites by how useful they are to a user’s search. The search giant doesn’t usually reveal its search statistics (even though many have attempted to estimate), but Google disclosed in 2012 that the year held 1.2 trillion searches across 146 languages. The majority of your website’s traffic most likely comes from Google, so you need to meet its ever-evolving set of ranking criteria. Sure, other search engines matter too, but considering the second-largest is the Google subsidiary YouTube, it’s safe to base your SEO strategies on Google’s standards. 

Take it slow

Your website won’t become an authoritative digital property overnight; it takes time and effort to master modern SEO and earn that clout. Traditional factors like quality backlinks, positive user engagement, optimized on-page elements, and stellar content still all contribute to your website’s credibility and rank over time, so don’t look for quick fixes. Stick the process and you will see results. 

Don’t worry if you haven’t been keeping up with recent algorithm changes. Even seasoned marketers can feel out of the SEO loop sometimes, especially when myths and speculations regarding algorithm updates spread online. These updates are real — Google announced a core algorithm update on May 4, after all — but you don’t need to worry. Just make sure you aren’t doing something that’s hurting rather than helping your brand’s website. Google will penalize you for spammy guest blogs, keyword-stuffed content, unnatural link building, and everything designed to unethically manipulate rankings. Simply put, Google wants you to earn your rank. 

With so much to focus on, how do you prioritize your efforts? Luckily, Google lays it all out for us: Focus on the user’s experience. Build trust and credibility with Google and the people visiting your site.

Becoming an online authority means prioritizing not only what’s important to you, but also what’s important to the people you want to reach. Here are three tactics you can use to support your modern SEO strategy this year: 

1. Create a thoughtful FAQ page

In a crisis, your strengths and weaknesses are spotlighted as a brand. People observe what you do and how you do it, and that will inevitably affect their opinions. This is especially true during the coronavirus pandemic. According to Edelman’s recent report, only 38% of respondents believe that businesses are doing well at putting people before profits. Your customers, clients, and consumers are humans, and they’ve craving comfort and connection right now. Prioritize the user experience with a frequently asked questions webpage. 

Everyone knows what an FAQ page is, but not everyone knows that it can be your best friend when it comes to optimizing your website for search. Having a thoughtful, helpful FAQ page shows that you care about your customers and search engines like businesses that care about their customers.

Plus, a robust FAQ page that’s marked up properly might receive a rich result on the search and action on the Google Assistant, dramatically expanding your reach. 

2. Organize content by topic

It is no surprise that internet usage is up. In fact, Comcast, the largest residential internet network in the U.S., has experienced a 285% increase in videoconferencing and VoIP use while people are quarantined at home. More people are streaming, searching, shopping, teaching, learning, meeting, and gaming online than ever before. This is your opportunity to meet the needs of users by organizing your content by topic.

When it comes to the amount of content displayed on a given webpage, don’t lose sleep over the word count. Adding value and being helpful doesn’t require a set number of words or characters. Just keep writing until you feel the content highly meets the needs of your users. You’ll end up with an article that Google will love.

Don’t pack tons of information across an array of topics onto a single page. Instead, focus on providing an exceptional user experience by ensuring individual pages are each dedicated to a unique topic. You’ll want to start this process by building out a content map to define the information architecture. As you gradually add pages and remove others, make sure your content map reflects these modifications. 

3. Get the tools you need to help with link building

When outreaching and creating content for link building and brand mentions, you’ll need to be helpful and human. It’s not just about hyperlinks — it’s about building relationships and providing value. During the coronavirus pandemic, let compassion guide this element of your strategy. Selectively reach out to domains with effective, modern content that audiences will find useful. There’s no better time to demonstrate your knowledge and support your audience. 

Link building is a critical part of the most effective SEO strategies. Just keep in mind that this is generally a long-term tactic, so you’ll have a hard time executing without some help. Luckily, there is plenty of software available to do the heavy lifting for you. The quickest and easiest place to start is Google’s own search console, which offers tools and reports to improve your website’s search presence. This service will show you which websites link to yours and confirm that Google can both find and crawl your site. 

As search engines learn more about users’ behaviors, they will continue to change. What was a widely accepted SEO best practice just a few years ago may now hurt more than help your goals.

The bottom line

You can’t predict what Google or any other search engine will do with its algorithm, but luckily, that’s not your job. Your job is to adapt, evolve, and do what’s best for your business and your customers.

Nick Chasinov is the founder and CEO of  Teknicks, a research-based internet marketing agency certified by Google in Analytics, Tag Manager, and a Google Premier AdWords partner.

When someone searching enters a query into a search engine, such as “top 10 movies of 2020,” the search engine may return results showing links to several webpages relevant to the search query.

It may also show results that link to several webpages that include lists of the top movies of 2020.

A patent granted to Google relates to lists of ranked entities in search results which are based on the documents that returned in response to queries about specific categories of entities.

This may be the top movies of 2020 or the best novels of 2020.

It could be the best science fiction books of 2020 or the best drama TV Series of 2020.

I searched for different types of entities which resulted in carousels showing off ranked entities for my queries:

“top movies of 2020”

“best novels of 2020”

“best science fiction books 2020”

“best drama tv series 2020”

The Process Behind Showing Ranked Lists of Entities

The method from the patent includes:

• Sending a query.
• Receiving information regarding documents relevant to the query.
• Identifying the entities associated with the documents.
• Determining a category for the query based on the query or a topic of documents returned and the entities within those documents.
• Determining that an entity list should be presented in response to the query.
• Presenting SERPs based on determining that the entity list should be presented in response to the query.

How Categories for Entities in Ranked Lists Are Determined

The SERPs may include a list with information identifying the entities.

Determining the category may include generating a score based on:

• Whether the query includes terms associated with the category, where determining the category may be based on the generated score.
• At least some of the documents are associated with a topic associated with the category, where determining the category may be based on the generated score.
• At least some of the entities, associated with the category, where determining the category may be based on the generated score.
• Whether the query includes blacklisted terms, where determining that the entity list should be presented in response to the query may be based on the generated score.

Ranking the Entities in These Lists

The process behind ranking entities may include:

• Where the search result document may include information about the entities in an order based on the ranking
• Generating a score for each of the entities, where the scores are based on the relevance of the particular entity to a particular document.

The process behind this patent involves:

• Receiving a query.
• Receiving information regarding documents relevant to the query.
• Identifying entities associated with the documents.
• Determining a category for the query based on the query, a topic of the documents, and the entities.
• Determining, based on the query and the category, that an entity list should be presented in response to the query.
• Presenting a search result based on determining that the entity list should be presented in response to the query.

The search result document may include a list with information identifying the entities.

The search result document may further include links to the documents that are relevant to the query.

This patent can be found at:

Generating ranked lists of entities
Inventors: Toshiaki Fujiki, Slaven Bilac, Kavi J. Goel, Shuhei Takahashi, Tomohiko Kimura
Assignee: Google LLC
US Patent: 10,691,702
Granted: June 23, 2020
Filed: August 31, 2017

Abstract

“A device may be configured to receive a query; receive information regarding documents that are relevant to the query; identify entities associated with the documents; determine a category for the query based on the query, a topic of the documents, and the entities; determine, based on the query and the category, that an entity list should be presented in response to the query; and present a search result document based on determining that the entity list should be presented in response to the query. The search result document may include a list with information identifying the entities.”

Extracting & Categorizing Entities

The patent provides examples of the extraction and categorization of entities from webpages, and other documents:

• A document may include text, images, etc., regarding entities.
• An entity may be extracted and/or identified from a document by comparing the text, images, etc., to a repository that includes information regarding entities.
• For example, an entity may be associated with the movie “Toy Story 3”.
• Another entity may be associated with the song “Party Rock Anthem”.
• Another entity may be associated with the book “Hitchhiker’s Guide to the Galaxy”.
• Entities may then be categorized. For example, categories of entities may include “movies,” “songs,” or “books”.

The repository mentioned in this patent is likely Google’s Knowledge Graph.

I searched for “Best Planet of the Apes Movies,” and if you hover over one of them, you see more information about each of the movies:

The patent does tell us that they may show additional information about attributes of ranked entities that they are returning:

“For example, user interface 145 may include images and links associated with the entities. Additionally, or alternatively, user interface 145 may also include other information associated with entities, such as attribute information–e.g., release date, music credit, producer credit, production company, or the like. As shown in FIG. 1C, user interface 145 may include, for instance, link 110 and image 115 associated with Toy Story 3, as well as links and images respectively associated with Inception and The Social Network.”

This is so that if you are potentially interested in finding out more about one of the entities shown in a ranked list, that you can find out enough to watch it, or read it, or listen to it.

Entity Extraction for Rankings

The documents that might be returned in response to a query could include documents such as “webpages, news articles, image results, blog documents, or the like.”

The results identification aspect of this process where entities are identified could supply information about the attributes of those entities.

Entities may be ranked based on factors to determine scores for those entities.

One may be if the entity is extracted from a document that has a less relevant topic:

“For example, assume that a first entity associated with the movie Toy Story 3 is extracted from a document with a topic relating to “football,” while a second entity associated with the movie Inception is extracted from a document with a topic relating to “movies.” Entity ranking engine 230 may determine that the first entity has a lower relevance to the result from which the first entity was extracted than the relevance of the second entity to the result from which the second entity was extracted.”

The terms in a document that an entity is extracted from may be reviewed.

An IR (information retrieval) score for the document from which the ranked entity was extracted may be used to determine an entity score used to rank that entity, based on how relevant the page was to the query.

So an entity to be ranked for “best science fiction novels of 2020” from a page that had a high IR score for the query, “best science fiction novels of 2020,” would potentially have a higher entity score than an entity from a page that has a high IR score for “Some OK Science Fiction novels of 2020.”

If a query for ranked entities might include a specific attribute, such as location published, and the query was something such as “Best American Science Fiction novels of 2020”, then ranked entities might be ones that identify the publication location for that entity.

The range of attributes used could be fairly broad, as described in the patent:

“Assume, for example, that the particular entity is associated with a ‘movies’ category. In this example, the attributes may include information, such as release date, lead actor(s) and/or lead actress(es), supporting actor(s) and/or actress(es), box office gross, executive producer credit, music credit, synopsis and/or summary associated with the movie, etc. In other examples, entities associated with different categories may be associated with a different set of attributes. Further assume that the particular entity is associated with Toy Story 3, a movie released in 2010 and that the query includes the term ‘2010.’ Thus, entity ranking engine 230 may identify that the query is associated with a release date attribute associated with the particular entity.”

Another factor may be if the query contains blacklisted terms.

These could be adult terms or offensive terms.

Another factor could be if the query is associated with entity lists. Entity lists include phrases such as:

• “top”
• “of 2020”
• “Best”
• “Best of”

If the query contains such a term, entities from pages that may also be relevant for those may score higher in a list of ranked entities.

When a query includes a trigger that calls for an entity list, it may show a carousel or a list of entities.

If it does not contain such a trigger, it may not show a list of entities.

Ranked Entities Results

This patent describes a process that you can easily get caught up in once you learn about it.

I found myself going through queries such as “Best Comedies 1975” and for years following that one.

And “Best TV Shows 1980” and for years after that.

“best TV shows 1980”

“best TV shows 1981”

“best TV shows 1982”

If you do those searches, you will see that the documents those entities come from in the rest of the SERPs for them are ones about “Top” or “Best” results, and the queries for them triggered query lists showing those carousels.

There are other ranked entities lists you can find as well, like Pulitzer award winners:

Or Best Houseplants for air quality:

Exploring these Carousels for ranked entities was almost as much fun as looking at the Semantic Ontology-based categories in image Search at Google, which I looked at in Google Image Search Labels Becoming More Semantic?

More Resources:

Image Credits

All screenshots taken by author, June 2020

Two Polands will battle it out in a presidential run-off next month — the often rural, older voters of the conservative incumbent against the younger city dwellers who mostly support Warsaw’s liberal mayor.

The current head of state Andrzej Duda, backed by the governing right-wing Law and Justice (PiS) party, emerged from round one on Sunday in pole position with 43.7 percent of the ballot.

Runner-up Rafal Trzaskowski got 30.3 percent, taking the Civic Platform (PO) member into round two on July 12, according to near final results from the national election commission.

Duda is expected to highlight the generous social benefits introduced on his watch as well as the Catholic values he holds dear, while Trzaskowski will probably stress his pro-democracy and pro-Europe programme.

Once the exit polls were out, the rivals — both 48 years old — lost no time in wooing those who backed the other nine candidates, as they climbed right into their campaign buses and set off to meet voters.

One poll by Ipsos showed that in round one Trzaskowski proved more popular than Duda among voters under the age of 29, as well as in cities with populations of over 200,000 people.

Duda meanwhile fared better than Trzaskowski in smaller-sized cities and towns, as well as in villages, where he won by a landslide. Duda also attracted more older voters, especially those 50 years and up.

– ‘Heavy heart’ –

“I think there’s a chance the current president will win in the run-off but he’ll have to work hard,” Michal, a 43-year-old lawyer, told AFP.

Fellow voter Urszula however said she was crushed by the results.

Presidential election in Poland

Vincent LEFAI, AFP

“It’s with a heavy heart that I note that if PiS voters rally, it’ll be tough,” the 60-year-old librarian told AFP.

“The people need to wake up. Maybe the heat and the coronavirus will rouse them!”

Duda and Trzaskowski will strive to attract those who backed the independent candidate Rafal Holownia, who came in third with 13.85 of the vote, and the far-right candidate Krzysztof Bosak, who got 6.75 percent of the ballot.

Analysts say many of those who voted for one of the also-rans did so in protest against Duda and the government, suggesting they could back Trzaskowski in the run-off.

The liberal candidate has “a lot of room for manoeuvre” in attracting voters from the others, said Maciej Onasz, a political analyst from the University of Lodz.

“I think it’ll be a brutal contest resorting to fear. Both parties will opt to mobilise voters in a negative way, against their opponent,” he said, quoted by the Gazeta Wyborcza daily.

– ‘Major confrontation’ –

During the campaign, Duda stoked controversy by echoing PiS attacks on gay rights and Western values, likening “LGBT ideology” to a new form of communism.

One analyst says he expects both candidates to campaign negatively during the run-off, attacking each other

MATEUSZ SLODKOWSKI, AFP

On Monday, Duda warned that coexistence between a liberal president and a conservative government would lead to political warfare and even early elections.

“It would mean a major confrontation between the government and the president, which is always bad for Poland,” Duda said.

“It would block all activity, leading to the overthrow of the government.”

For Trzaskowski, the problem is the conservatives’ tight grip on power.

“What hurts us is this government’s monopoly on power. They have politicised every institution, every oversight institution, in such a way that it is hard now to hold them accountable,” Trzaskowski said on Monday.

“The government and the president believe that the people are there to serve those in power, when in fact it’s those in power who should be serving the people.”

Were Trzaskowski to win, it would spell trouble for the PiS government, which has run afoul of critics at home and abroad for judicial reforms that the EU says threaten the rule of law.

Microsoft’s Digital Marketing Center.

Microsoft’s Digital Marketing Center is moving past pilot stage, the company announced Monday. The platform is designed to help small and medium-sized businesses manage digital campaigns across multiple networks from one interface.

Digital Marketing Center enables small businesses to manage organic social media and paid search and social campaigns across Microsoft, Google, Facebook and Instagram, and Twitter’s ad platforms. The product came out of the company’s experimental project lab, Microsoft Garage.

A fully AI-powered approach. Similar to Google Smart campaigns, which aim to simplify campaign set up and management for SMBs, Digital Marketing Center uses Microsoft AI to power ad keyword and audience targeting and bidding. Advertisers set their goals, location targeting and budgets.

Unlike Smart campaigns, in Digital Marketing Center, advertisers can choose to build their own ads, use automated ad copy or modify auto-suggestions.

Budgets are automatically optimized across channels and platforms.

Organic social management. Businesses can also create, schedule and publish posts on Facebook, Instagram and Twitter for up to 10 profiles from the Social dashboard.

From the Social “Inbox” shown below, you can reply, like and direct message your users on Facebook, Instagram and Twitter.

The Reports section shows impressions, engagements and audience growth across these channels.

Why we care. This is a very interesting product move by Microsoft in thinking beyond its own turf. It could help solve a major pain point for SMBs. There are platforms like HubSpot already positioned as all-in-one marketing centers for SMBs, and certainly Microsoft’s Digital Marketing Center doesn’t have all the CRM features of a HubSpot but it’s free and its simplicity may be a real draw for overwhelmed, resource-stretched SMBs.

It could also, of course, boost adoption of Microsoft Advertising by SMBs using the product. Microsoft could eventually charge for the product if it catches on (the release states it’s “currently free” to use).

If you’re interested in trialing it, start here. The beta is open to U.S. business only at this point.

This story first appeared on Search Engine Land.

About The Author

Ginny Marvin is Third Door Media’s Editor-in-Chief, running the day to day editorial operations across all publications and overseeing paid media coverage. Ginny Marvin writes about paid digital advertising and analytics news and trends for Search Engine Land, Marketing Land and MarTech Today. With more than 15 years of marketing experience, Ginny has held both in-house and agency management positions. She can be found on Twitter as @ginnymarvin.

Chris Sherman has been serving the SEO and SEM industry since the super early days, from Search Engine Watch and Search Engine Strategies to Search Engine Land and Search Marketing Expo. He was there from the beginning and in 2019, at the last SMX event that year, Chris announced he is retiring. Here is my interview with this man, a man who helped form this space as we know it.

His first job in SEO was working for The Mining Company, which eventually became About.com (and now Dotdash). This site launched in 1997, just so you know the age of it. His job there was to write articles about the state of search, covering search engines like Excite.com and other older search engines. He met up with Danny Sullivan when Danny did the keynote and the Internet Librarian Conference. Danny spoke about his early paper on the Webmasters Guide to search engines, and this was the first of its kind. Then in 2001, Danny sold Search Engine Watch to Internet.com, and then he was hired to work at Search Engine Watch under Danny. Chris Elwell was the person who was involved in the acquisition and they moved into Search Engine Strategies, the conference.

Then that company sold Search Engine Watch to Incisive Media, he stayed on for a bit when Danny (and others) went to start Search Engine Land. But he joined Search Engine Land and the Third Door Media company shortly later to help form that company. Both companies he worked at became authorities in the space.

He is going to be working on personal projects; word-working and cooking. He said he isn’t sure if he is done professionally but he has times to think.

You can follow Chris Sherman on LinkedIn.

You can subscribe to our YouTube channel by clicking here so you don’t miss the next vlog where I interviews. I do have a nice lineup of interviews scheduled with SEOs and SEMS, many of which you don’t want to miss – and I promise to continue to make these vlogs better over time. If you want to be interviewed, please fill out this form with your details.

Forum discussion at YouTube.

Snapchat users can now use AR tech to virtually try on Gucci footwear to see if they like the merchandise and then have the option to buy it directly from that location. Image courtesy of Gucci

Fédération de la Haute Couture et de la Mode is digitizing key events as insurance against future health outbreaks and also to keep up with the needs of the times and current market behavior. Image credit: Fédération de la Haute Couture et de la Mode

Doctors, nurses and other hospital staff staged protests across France Tuesday to press for pay hikes and budget increases for a healthcare system that was pushed to the edge by the coronavirus pandemic.

The rallies came as the government prepares to wrap up on Friday weeks of talks with health workers on hospital overhauls in response to the crisis.

Employees have long complained about insufficient staff and low pay that prompts doctors and nurses to take jobs at private clinics instead. That issue led to a series of strikes over the past year.

Prime Minister Edouard Philippe promised “significant” pay increases when kicking off the talks last month, and officials have already put an additional 6.3 billion euros ($7 billion) on the table.

But that amount falls far short for many employees. They say French healthcare workers are already among the lowest paid among the OECD group of developed economies, because of years of budget cuts.

“They’ve promised six billion euros but I’m still waiting to see,” said Louis Rios, a psychiatric nurse in the Essonne department south of Paris, during a march in the French capital.

“We want a significant gesture, and now — not in three years,” he said.

The SUD-Sante union chief, Jean-Marc Devauchelle, has called on the government to boost take-home pay across the sector by 300 euros a month, a move that would cost some 14 billion euros.

“We need acts that live up to the gratitude,” Philippe Martinez of the CGT union said at the Paris demonstration — a reference to the nightly rounds of applause for hospital staff during the height of the coronavirus crisis.

President Emmanuel Macron, who has made social justice a key theme for the final two years of his term, is expected to announce measures resulting from the talks as soon as next week.

The work of a search engine optimization (SEO) consultant revolves around one central theme:

Data.

Especially keyword data.

We collect it from a variety of third- and second-party sources, perhaps even via self-made tracking tools, to then start crunching the numbers and eventually delivering valuable insights to our bosses, clients, or prospects.

However, only running a few tools and employing some analytical magic is not going to cut it.

We also need to be thoughtful about how we interpret data from keyword tools and deal with any inaccuracies or inconsistencies.

Just like any software program, each keyword tool has a characteristic mechanism in place for collecting, aggregating, and manipulating data.

Similarly, tools’ workings affect how they handle queries and present the output keyword data.

An essential part of a marketer’s job function is to validate whether the data values stored for these keywords are represented in a consistent and unambiguous form.

Meaning, is the keyword data I am working with accurate?

The simple answer:

No.

Comparing data values of different tool providers for a set of keywords already proves to contain large inconsistencies – not only in data values but also in if and how your output data is presented.

This study, by my company, OAK, attempts to find clarity by exploring data accuracy and reliability with respect to second- and third-party keyword tool data.

Specifically, this study examines the following topics:

• Data collection: How do keyword tools collect their data?
• Data handling: How do keyword tools manipulate data?
• Data validation: Validating keyword data values.
• Role as an SEO Consultant.

The primary purpose of this study is to grow awareness about the complexity surrounding keyword data values and tool providers’ data collecting and processing mechanisms.

Google Search Console

Let us begin at the beginning: Google Search Console.

It is a second-party tool from Google that collects behavioral data for a single domain or entity and, after manipulation, injects the data into the front-end interface.

The mere fact that Google collects and processes the data might you wonder: how close to reality are the data values of the projected data?

This question poses an immediate challenge: Search Console data is not 100% validatable.

Luckily, Google is, to a point, transparent and provides various explanations for why your data values do not reflect reality or add up as you could expect.

A few of them are:

• To protect the privacy of the user. The click is sometimes not credited to the search term. Search Console, however, does register the click, causing discrepancies between the table and diagram data.
  • The same can apply to branded queries.
• Clicks could come from bots.
• In some cases, selecting certain filter combinations can also lead to differences between the diagram and table data.

Unfortunately, only the G-giant has access to the exact data values, which means verifying the accuracy of Search Console data is a troublesome process.

The reliability of keyword data increases, however, with third-party tools.

These are tools like SEMrush, Ahrefs, Keywordtool.io, Searchvolume.io, and many others.

To find answers, this study explores the mechanics of these keyword tools apply.

Unfortunately, the companies running these tools disclose little to no information about how they collect, aggregate, or manipulate their data.

It seems fair.

A chef doesn’t just give away her or his world-famous recipe. Hence we attempt to generate insights with the help of the following approaches:

• Using and comparing the tools.
• Inquiring at the customer service departments.
• Reading the FAQ sections and utility pages.

1. Data Collection: How Do Keyword Tools Collect Their Data?

In general, there are five kinds of resources through which keyword tools accumulate their data:

Google Ads API / Keyword Planner

Keyword data is gathered directly from Google’s keyword database through the Google Ads API.

As is the case with Search Console, Google Ads first manipulates the data before injecting it into the database.

Clickstream Data by Aggregators & Data Brokers

Clickstream is nothing more than data derived from consumers’ online surfing behavior.

Aggregators gather this data in a variety of ways.

Large, until recently active aggregators, were, for example, Jumpshot or Hitwise.

Wherefrom do they get their data?

• Browser extensions and plugins
  • A homemade plugin or extension of the aggregator itself.
  • They pay external third party browser plugins to share consumer data
• They pay internet service providers for access to the data in an “anonymized” data feed.

The aggregators then sell the data to keyword tools such as Ahrefs, SEMrush, and Moz, among others.

Browser Extension & Plugins

Keyword tools can also directly acquire consumer data from external third party browser plugins.

Numerous browser extensions have been developed to help marketers.

Despite the nifty functionalities, these plugins and browser tools often pursue shady practices.

Providing consent before using an extension is common, but we usually have limited knowledge about the practices to which we give consent.

By consenting, you can permit these tools to:

• Collect your online surfing behavior.
• Retrieve client data from Google Analytics, Search Console, or other tracking software.

And most worrisome:

• Share the data with third parties such as aggregators or keyword tools.

These browser extensions might have access to any potentially sensitive data and are ordinarily not compliant with your customer or business’s GDPR.

Exercising care when working with extensions is necessary to guarantee data security.

Some keyword tools have also developed their own browser plugin or extension.

Moz, for instance, introduced MozBar, an all-in-one extension with all kinds of useful features.

Browser plugins developed by the established keyword tools presumably wouldn’t pursue any malicious practices, but they can collect online behavior and use it to attune their data values.

External Tools

Keyword tools additionally retrieve data via APIs from external parties that collect online surfing behavior data, such as GrepWords in the past.

Keywordtool.io, for instance, obtains data from Keyword Planner but also other third sources. They disclose some advice, albeit rather general:

“Keyword Tool provides an external API that gives you the keyword suggestions that you would never be able to find in Google Ads. Keyword Tool uses autocomplete data while Google Ads hides valuable keywords that could be found using autocomplete.”

Own Tools

Some keyword tools instead have their own programs or software set up to accumulate keyword data.

These five methods of collecting and examples only unveil the tip of the iceberg.

There exists a vast web of companies and tools within this world of collecting, exchanging and selling keyword data.

At least a bit better understanding of the picture will undoubtedly help us to realize that keyword data displayed on our SEO tools, Google Sheets and dashboards are barely more than a product of an ambiguous construct.

To Sum Up

Keyword tools collect data from five different types of sources.

It’s common to use multiple data sources from different data source types.

2. Data Handling: How Do Keyword Tools Manipulate Data?

The next step in adopting a more critical stance toward keyword data is by learning how tools aggregate and manipulate the data they’ve obtained.

It is close to impossible to find out just how exactly tools run this procedure.

The practice of getting this proprietary information is equal to attempting to discover Coca Cola’s recipe – futile.

Instead, let us settle with the notion that aggregation in itself may further taint the accuracy and reliability of data.

One could argue the reverse that aggregating data can automatically flatten out any extreme data values.

After all, merging these data sources into a single coherent aggregated type will yield a better approximation of the average metric values.

Even though it’s a legitimate stance, keyword tools nevertheless gather data from resources that in itself may be incalculable, biased, and incorrect in their measuring mechanics.

Regardless, keyword tools do more than mixing different data sources into a single set of data.

Schematic representation of keyword tools’ workings in collecting, manipulating and querying data.

Running the Data Through an Algorithm

Some tools have developed an algorithm that functions as a filter for their collected data.

For instance, SEMrush explains:

“To ensure the highest level of accuracy, SEMrush uses its Neural Network – a combined algorithm that references various sources of data and recognizes patterns in the same way the human brain understands patterns. The data sources in our network include clickstream data in addition to our own database of backlinks and organic search engine positions. “

It seems a logical explanation that SEMrush uses its algorithm to validate the obtained external data and attune the aggregated metric values where needed.

(un)Grouping

Keyword tools can group keyword data, which proceeds in two different ways:

• They group metric values of search term variations into one.
• They group variations of search terms into one.

This grouping mechanism relies on four linguistic determinants:

• Plural vs. singular nouns in the keyword.
• Combinations of articles and prepositions.
• Usage of regular, comparative, and superlative adjectives.
• Placement of adjectives or interrogative pronouns.

Volume Grouping

Let’s begin with an example.

We have two different search terms, “door handles” and “door handle.”

Some tools, whether keyword tools, aggregators, or other data collecting tools, merge the individual volume values into one aggregate total, and display this total for both keywords.

For example, keywordtool.io groups the keyword, whereas searchvolume.io does not.

That is what it looks like for the U.S.:

Two things stand out immediately:

• Keywordtool.io attributes the same volume value to both keywords (plural and singular), whereas searchvolume.io does not.
• Volume values of Searchvolume.io are significantly lower compared to that of keywordtool.io.

Let’s also take a look at Ahrefs.

Ahrefs collects its data from Keyword Planner, among other sources.

According to the customer service department, Ahrefs ungroups the keywords that Keyword Planner groups together.

The following table is similar to the previous one, but this time we have included data from Ahrefs and pulled the same query for another country.

Two things stand out:

• Ahrefs data values reveal different volume values compared to both keywordtool.io and searchvolume.io.
• For the Netherlands, Ahrefs does assign a higher value to the singular form “door handle” compared to the plural type “door handles.” It’s diametrically opposed to Searchvolume.io’s values.
  • However, for the U.S., Searchvolume.io and Ahrefs exhibit the same equilateral distribution.

Query attempts with other keyword sets give us similar results. In some cases, tool X presents the largest values, in other cases, tool Y or Z.

One thing is sure: data values are scattered, questioning the reliability of data values.

The next table lists a set of popular keyword tools and whether or not they group keyword volumes:

Keyword Grouping

Besides volume clustering, the grouping effect also applies to search terms.

The first linguistic determinant in keyword grouping is singular vs. plural usage of nouns.

Singular vs. Plural

Keyword tools can group nouns into either the singular or plural form.

However, this doesn’t necessarily mean the other version, whether singular or plural, doesn’t exist in the tool’s database.

Tools choose which form to display in the output.

We’ll illustrate with Keyword Planner.

Let’s suppose we want to retrieve the U.S. search volume of the next four keywords.

Next, we choose the tab Historical Metrics showing the following data table:

Two things that stand out immediately:

• Keyword Planner returns only data for two out of four keywords.
• Keyword Planner returns only the singular form of the nouns.

When repeating this query for other countries, we can’t observe any logical pattern between the query and the presented data.

For example, for the Netherlands, Keyword Planner serves the following table:

For those with limited comprehension of Dutch language:

• “Deurklinken” (i.e., “door handles”) is plural.
• “Deurpost” (i.e., “doorframe”) is singular.

Keyword Planner thus groups based on volumes as well as keywords.

Looking at both the U.S. and the Netherlands, we can infer that Keyword Planner’s database does contain the data values for both the singular and plural form.

To be sure, let’s rerun the query.

Only this time for just the U.S. and with the plural forms of the nouns:

• “door handles”
• “doorframes”

The results:

While Keyword Planner omits either the plural or singular noun in the export, its database does include data on all four keywords.

Also, the keyword volume values reflect the aggregated volumes for both the plural and singular form.

Likewise, there seems no clear ground on which Keyword Planner decides which form to display other than an arbitrary one.

One can further explore this topic by, for instance, comparing multiple countries, industries, volume ranges, and languages.

More in-depth exploration, however, is beyond the scope of this study.

The fact still is, it causes a great deal of confusion.

Combinations of Articles & Prepositions

Tools group keywords in instances where search terms comprise articles and / or prepositions.

To illustrate, we provide an example of Keywordtool.io.

We compiled a list of eight keywords to extract from Keywordtool.io’s database:

• “legislation in the united states”
• “legislation in united states”
• “legislation the united states”
• “legislation united states”
• “legislation in the us”
• “legislation in us”
• “legislation the us”
• “legislation us”

For the conspicuous reader, the list is as follows:

• We used two ways to write down the United States: “United States” or “US”.
• We created four combinations of the article “the” and the preposition “in”.
  • “In the”
  • “The”
  • “In”
  • – (So neither “the” nor “in”)

Querying the data from Keywordtool.io’s database gives us the following data:

Several things stand out:

• The query prompts results for only five out of the 10 keywords. There seems to be no obvious factor that decisively affects this particular output. The excluded keywords:
  • “legislation in united states”
  • “legislation the united states”
  • “legislation the us”
• Keywords with both the written form “United States” and the abbreviation “US” are presented, but, clearly not for the same variation of articles and prepositions:
  • Listed: “legislation in us”
  • Not listed: “legislation in united states”
• The combination without both the article and preposition is given for both the “US” and the “United States” variant. Still, both display other volume values:
  • “legislation united states” — 210
  • “legislation us” — 40
• Grouping of volumes occurs “cross-keyword”. Both variants of “US” and “United States” as well as preposition and article variants exhibit the same metric values. It means keywordtool.io groups the volume values of the following keywords:
  • “legislation in us”
  • “legislation us”
  • “legislation in the united states”
  • “legislation in the us”

Pertinent questions that come to mind:

• Why is it that the combination “legislation united states” eludes clustering?
• To what extent do articles and prepositions play a part in keyword grouping?
• How come the particular four ungrouped keywords show clustered volume values?
• Is there any precise, explicit mechanism in place that regulates the presentation of queried data?

These are legitimate questions to which we, unfortunately, do not have a grounded answer.

Keywordtool.io uses the Google Ads API to retrieve keyword data.

Can we then also expect the same to happen with Keyword Planner?

We tested it by running the same query for Keyword Planner:

It produces quite a different scenario.

Besides the previous observations, we can also observe that Keyword Planner only lists four out of the eight keywords.

Also, values are given only for two out of the four keywords.

Usage of Regular, Comparative & Superlative Adjectives

Adjectives or interrogative pronouns and comparative and superlative adjectives do play a part in tools’ grouping mechanisms.

At first, it might not seem such a big deal. For instance, if we search for “clean hotels London” or “cleanest hotels London”, the intent and the corresponding SERP results are both pretty similar.

In other cases, however, visitors’ needs and intentions do profoundly differ. Let’s consider the next three keywords:

• “low blood pressure” – I have low blood pressure and would like to have information on blood pressure levels that are considered low, and perhaps what to do about it.
• “lower blood pressure” – I have high blood pressure and I would like to have information on how I can lower my blood pressure levels.
• “lowest blood pressure” – I probably feel quite bad, and I would like to know what blood pressure levels one can have without it being life-threatening.

Looking at the example, we can see differences in:

• Audience.
• Audience’s health conditions.
• Informational needs.

This example highlights the obvious fact that we shouldn’t address these differences with the same content, or cluster these three keywords into one topic bucket.

What happens if we pull data from keyword tools for these particular search terms?

Inserting them into Keywordtool.io prompts the following results for the UK:

It’s pretty clear: all metrics presented share the same value.

Before drawing any conclusions, let us first cover the final determinant.

Placement of Adjectives or Interrogative Pronouns

The variable placement of adjectives or interrogative pronouns constitutes the last linguistic determinant in the grouping mechanism.

It doesn’t happen too often, but sometimes we place adjectives or interrogative pronouns at the middle or end of a phrase instead of the beginning.

For instance, one can search for:

• “electric scooter fast” or “fast electric scooter”
• “how fast electric scooter” or “electric scooter how fast”

Either case carries the same need for information.

It becomes rather interesting when we add comparative or superlative adjectives to these examples and create new combinations such as “electric scooter faster”.

The point is, differences in interrogative pronouns or comparative and superlative adjectives can exhibit divergences in users’ intent and needs, and the type of audience users belong to, as the next table illustrates:

Unfortunately, such latent information is hard to derive from these third-party tools’ query output, especially when grouping is at play.

The next table by keywordtool.io illustrates this:

Perhaps not surprisingly, the volume reflects aggregated values.

The conspicuous reader notices that the table merely lists keywords in the singular form.

Converting singular to plural form gives us the next data provided by keywordtool.io:

For data accuracy and reliability purposes, the immediate action here is to validate, to the extent possible, the volume values attributed to each keyword.

One way to do this is by querying the same four keywords in other keyword tools.

Other tools prompt different results. For example, SEMrush shows no results and Keyword Planner was similar to Keywordtool.io.

Ahrefs and Searchvolume.io did present data for all variations and, even more interesting, with disproportionately smaller volume values.

For instance, the query for the UK in Searchvolume.io prompts this data table:

That is a staggering 70 times the difference of 310.

It’s true that the volume value of 3.600 already reflects the aggregated volume for the set of six keywords.

But sadly it happens all too often that marketers record all six – or perhaps even more – variations in keyword analyses.

We can propound the idea of choosing one variation and omitting the other combinations. But it will not solve the issue.

The embedded information in keyword variations about users’ intents and needs can simply differ, and thus any form is relevant to include.

Imagine overlooking such a detail, failing to see that each keyword potentially belongs to different clusters of keywords destined for different pages.

And basing your traffic and financial projections on these numbers.

It’s a compelling image, albeit a little amateurish. Nonetheless, something that happens frequently.

There is one observation left unattended.

Keywordtool.io presents disproportionately larger volume values as compared to that of, for instance, Ahrefs.

Despite them both retrieving keyword data from Keyword Planner.

Apart from the tools’ data handling systems, what could cause such a difference?

Spelling Errors

Keyword tools differ in how they deal with spelling errors.

Some, like Searchvolume.io, leave out any spelling error variant in your query output.

Others, like Ahrefs and Keywordtool.io, do include spelling errors variants.

They both show the data values for every single keyword in your query as long as the keyword’s correct spelling variation exists in its database.

But as Ahrefs ungroups keyword data originating from Google Ads API, it does attribute unique metric values to each spelling error variant.

Keywordtool.io, on the other hand, adopts the grouped keywords and metric values it retrieves from Keyword Planner, causing all spelling error variants to show identical metrics.

Misspellings often occur with brand names easily prone to being misspelled.

Think of brand names such as Audemars Piguet, Breguet, Douwe Egberts, Schwarzkopf.

Let’s take a look at “Douwe Egberts”.

I am Dutch, and as a native, I am familiar with the varieties of errors one can make.

For instance:

• Is Douwe with ou or au?
• Is Egberts with g or ch or even with gh?
• Is it Egbert or Egberts?

Point is: what happens when we query a list of one single keyword misspelled in 26 different ways?

Despite the grouping mechanisms in place, Keywordtool.io serves you every unique misspelled keyword combination:

Ahrefs’ situation is a bit different. The query output is as follows:

Things that stand out:

• The query excludes 6 out of the 26 keywords.
• Ahrefs seems to independently attribute data values per metric.
  • Except for the first result, which is the correct spelling variation, all other keyword variations have either one or multiple metric data values missing.

Keep in mind that it only works with the ‘list’ mode. The ‘explore’ function will only serve the correct spelling variation.

Omitting ‘PPP’ Data

Google Ads API omits keyword data involving ‘PPP’ topics.

It means that other tools retrieving data from Keyword Planner also face this limitation unless they enrich their database with data coming from other sources.

For security purposes, Google disallows keyword tools to retrieve keyword data revolving around Porn, Pills, and Poker.

Think of keywords like “cannabis” or “full house” but also keywords such as “Koffiemachine huren”.

“Huren” is Dutch for renting. But it also means “whores” in German.

While this ruling is not a matter of direct manipulation, it does complicate marketers’ data collection and analyses.

A selection of keyword tools and whether they provide”PPP”-data for your query:

The above examples illustrate the chaotic nature of keyword tool mechanisms and the hazards they impose on SEO consultants’ work.

To Sum Up

• Keyword tools do not necessarily show all keyword variations and corresponding metric values.
  • Potential determinants: Tool’s functionality, safety or security measures, or missing data in the database.
• To our knowledge as outsiders, it seems the particular display of combinations of keyword and metric value variations is randomly “chosen”.
• Grouping applies to both the numerical values and search terms.
• Linguistic determinants for keyword grouping:
  • Plural vs. singular.
  • Usage or non-usage of articles and prepositions.
  • Placement of adjectives or interrogative pronouns.
  • Usage of comparative and superlative adjective.
• Grouping occurs both within a particular group and across group variations.
• Grouping occurs at random.

3. Data Validation: Validating Keyword Data Values

Keyword data validation is possible, yet without access to uncorrupted data, it becomes an act of finding the closest approximation to the keyword’s actual data values.

One option is to benchmark keyword impression data values from Search Console to volume values of third-party keyword tools.

Search Console data isn’t 100% reliable either, but it is as close as we can get.

Early in 2020, we designed a study to determine the accuracy of keyword data with a set of 160 keywords from a variety of industries.

The study tackled these two questions:

• For each keyword tool, what is the average deviation % of keyword volume values for the whole set of keywords compared to Search Console impression data values?
• For each keyword tool, what is the variance of all deviation % for the whole set of keywords?

The former gives us insights into the degree of accuracy for any given keyword’s volume value.

The latter question determines to what extent the deviation % of each keyword is spread out from the average deviation value.

As values can both deviate negatively and positively, it does not suffice to merely show the average deviation %.

It is the combination, however, of both scores that yield the best results in determining the data values’ accuracy and reliability.

As these visualizations illustrate, we see that exclusively looking at either variance or average deviations can prompt erroneous representation of the situation:

Measuring the variance of the average deviation % enables us to determine the scatteredness of each keyword’s deviation percentage.

Large dispersions hints to lower accuracy and thus reliability of the keyword volume values.

To put it differently, the more substantial the variance, the higher the likelihood of selecting a keyword from the data set showcasing a more inaccurate volume value than the average volume value deviation.

These were the key findings:

• Twinwords volume data showed the largest positive average deviation to Search Console impressions: +37.13%.
• Searchvolume.io volume data showed the largest negative average deviation to Search Console impressions: -34.71%.

• The frontrunners with the largest variance
  • Twinwords: 5,259
  • Keywordtool.io: 5,256
  • Keyword Planner: 5,188
• The frontrunners with the smallest variance
  • Serpstat: 0.124
  • Searchvolume.io: 0.149
  • Ahrefs: 0.153

Ideally, tool providers exhibit numbers close to zero for both the average deviation and the variance of the average deviation.

These findings show otherwise. Specifically:

• The grouping effect drives the large variance and average deviation score by the frontrunners.
  • Both Keywordtool.io and Twinword get their data directly from Google. And since Google Ads applies grouping to keyword and data values, Twinword and Keywordtool.io automatically adopt this effect.
• Keywords with the largest data value deviations also appeared to be keywords with grouped data values.
• Serpstat, Ahrefs, and Searchvolume.io present variance numbers close to zero. These tools do not apply any clustering.
• Serpstat and Searchvolume.io show considerably lower average deviations. It suggests that volume data is on average lower than what you would expect according to the Search Console.
• Although Searchmetrics’ keyword volume values barely deviate on average to Search Console’s impression values, the individual data values are further removed from the mean, suggesting a higher degree of inconsistency in keyword data values.
• The numbers of Ahrefs and KWFinder exhibit the closest approximation to the keyword’s actual data values.

Data values from third-party keyword tools vary widely and seem to fail in providing unambiguity or consistency.

The findings further give the plausibility to the idea that mechanisms in handling data queries, and collecting or manipulating data, can add to delivering erroneous keyword data.

Can We Then Validate the Accuracy of Keyword Data in Another Way?

Together with Sander Tamaëla, a Dutch Freelance SEO-expert, we came up with a way to validate the accuracy of third-party keyword volume values with the help of Google Search Console and Google Trends data.

The idea was as follows:

• We picked one noun and selected both the plural and singular form.
  • We had validated the accuracy of search volumes with GSC monthly average impression values.
• Then we retrieved the volume data from two or three random keyword tools.
• We then placed these two keywords in Google Trends.

With this setup, we could determine the relative interest between the two keywords.

Our assumption here was that Google Trends’s relative interest scores reflect the purest data values.

As such, the relative interest score should reflect a ratio similar to that of impression values in Search Console.

• Next, we expanded the set with keywords – for which we know we have an accurate approximation of the impression value – from several volume ranges.
• Then we compiled a training set.

The idea was to determine per volume range the deviation per keyword volume value based on the relative interest scores of Google Trends.

Unfortunately, things didn’t work out as planned.

After we had challenged the assumption that Google Trends data depicts accurate values, we discovered that Google Trends isn’t entirely reliable either.

To test the reliability of Google Trends we set up the next test:

• We selected five keywords with very similar monthly impression values in Search Console.
• We then added these five keywords in Google Trends.
  • We made sure that we had chosen the same 12 month period for Search Console as for Google Trends: 1 December 2018 until 30 November 2019.

One of the five-keyword sets, in Dutch:

The next chart illustrates each keyword’s impression value deviation from the mean:

The monthly impressions average deviation in percentages was only 1.92%.

Unfortunately, it is only possible to select up to five keywords in Google Trends, limiting our sample to five keywords.

Such a sample size is statistically not a significant representation of the population. The only alternative was to repeat the test setup with different sets of keywords.

If Google Trends is reliable, we would have expected that the average interest ratio between the keywords in Google Trends is virtually the same.

What was the result?

For those five keywords, we observed ratio levels of relative interest score that were varying disproportionally:

Google Trends’ average interest scores:

The ratio of three out of five corresponds to the ratio of Search Console impressions.

But the remaining two keywords differ significantly, with an average deviation in percentages of 31,57%.

Again, with a sample size of five, the average deviation output is not significant.

But by repeatedly testing the setup for different keyword sets, we observed a similar pattern.

Two other examples of Google Trends’ relative interest scores for five-keyword sets:

To put in perspective, the ratio average deviation percentages of Search Console are respectively 2.73% and 1.62%.

Google Trends’ average deviations thus show significantly larger percentages than those for Search Console’s impression value ratios.

Can we then draw any conclusions here?

As outlined in the beginning, Search Console isn’t always showing the most accurate representation of reality.

However, the designed setup to test Google Trends’ data accuracy and reliability provided evidence suggesting that data from Google Trends isn’t consistent or unambiguous either.

Does this mean that we can no longer use these tools? Or perhaps only some?

Not necessarily.

But, it doesn’t hurt to be aware of the demerits from keyword tools.

4. Role as an SEO Consultant

The primary purpose of this study is to grow awareness about the complexity surrounding the data values of keyword tools.

The next step after awareness is to incorporate critical thinking allowing us to recognize any faulty habits.

Common pitfalls to avoid:

• Taking volume data values as granted.
• Merging keyword volumes from multiple tools without further checks.
• Skipping the spellings check.
• Ignoring the grouping effect or not validating groupings.
• Inferring hard conclusions from your keyword volume data calculations.
• Not providing a reliability clause for your findings in the communication to your customer or prospects.

We cannot afford to take data from keyword tools for granted.

To build expertise and provide solid, reliable advice, we ought to set standards for how we work with keyword data.

How Will This Impact Your Role as an SEO Consultant?

I would argue that it starts with establishing a greater sense of accountability.

Remember the earlier example of overlooking a minor detail?

Imagine that happens.

You give this killer PowerPoint presentation. The prospects at the table are completely baffled by your story; you just landed a new client!

A few months pass by, and you discover that the total volume amount of your keyword data set is only 60% of the total amount you initially communicated to your client.

Assuming your analysis included virtually all existing keywords relevant to the business, such a mistake is difficult, perhaps even impossible to rectify.

Especially if your client’s case is specific to a niche or product cluster, you simply won’t find other relevant keywords to close the volume gap.

To avoid such disasters we recommend to incorporate the next worthwhile practices:

• Spend more time on your data analysis. A correctly performed keyword analysis takes time. Quality ≠ quantity.
• Validate your keyword data values.
• Double-check your data for irregularities
• Have your ‘facts’ straight.
• Do you have to make a presentation and draw conclusions? Make sure you at least have a proper contextual story ready to support your claims.

Your boss or client might not understand why individual efforts during the analysis have to take a substantial amount of time.

Be open and transparent to clients and prospects about the required efforts to ensure the continuous delivery of quality. It creates trust and fosters mutual bonding.

Telling your client beforehand is thus indisputably better than explaining your mistake afterward.

That will irreversibly compromise the relationship with your client.

Final Notes

• This study’s goal is not to place keyword tools in a bad light.
• Neither do I argue that keyword tools are in any way deficient. The reason I have provided the examples is purely to evoke a sense of awareness surrounding the accuracy and reliability of keyword data.
• This study did not include other search engines such as Bing, Yandex, and Yahoo.

Appendix

The study’s setup was as follows.

We selected a set of 160 keywords from various Google Search Console accounts. The selection of keywords depended on whether all of the following conditions were satisfied:

• The keyword must have had a top 3 SERP ranking for 12 consecutive months without any temporary dips reaching lower rankings.
  • This facilitates an as close as possible approximation of the real average monthly impressions count, based on a 12 month period.
• The keyword’s monthly impression count is 1000 or higher.
  • This increases the likelihood that each participating keyword tool’s database contains data on the selected keyword (long-tail keywords are less likely to be registered in keyword tool databases).
• The keyword should not be subject to seasonality.
  • It increases the likelihood of consistent top SERP rankings throughout the year.
• We also made sure that the 12 month period of GSC data matched the 12-month period with which keyword tools calculate their monthly averages.

These criteria were set in order to establish an accurate recording of calculated monthly impression values.

Most keyword tools calculate their monthly average volumes in a similar vein.

More Resources:

Image Credits

Featured Image: Created by author, June 2020
Infographic: Created by author, May, 2020
Screenshots taken by author, April & May, 2020

There was a time—and not terribly long ago—that getting marketing messages to a mass audience was relatively easy: identify a target demographic and then figure out the mix of broadcast and print ads that balanced reach and repetition within a given budget. In those days, media and entertainment choices were relatively limited. Even when people had many cable channels to choose from, the available content was offered by one provider who managed a relatively captive audience.

Today, marketers look at this era as a quaint relic of the past, much the same way they might with the world of Mad Men. With the rise of cable-cutting (swelling by more than 4.5 million US households in 2019 alone) and a growing generation of “cord nevers,” it’s clear the era of traditional mass messaging is over. Interestingly, despite this seismic shift in consumer preferences for providers, the television remains the screen of choice for many viewers.

With that being the case, the challenge becomes: how can we transition from mass messaging to targeted storytelling? The answer lies in fundamentally rethinking strategies using a more nuanced palette of options, capitalizing on the flexibility and power offered by connected, “smart” TVs. 

Does your story need to extend its reach beyond a traditional TV audience? Is your brand’s audience more likely to be found watching Hulu, Netflix, or Disney+? Do you want your messaging to be aligned with specific kinds of content regardless of device? The technology and measurement infrastructure now exists to serve any—or all—of these needs.

Best capitalizing on these opportunities requires a rethinking of the art of storytelling—reimagining the possibilities of connecting and resonating with audiences in new and compelling ways. It’s an exciting creative challenge, and the brands that rise to the occasion are the ones that will forge a new kind of branded content for the future—all the while striving to target an “audience of one.”

Opinions expressed in this article are those of the guest author and not necessarily Marketing Land. Staff authors are listed here.

About The Author

As Chief Innovation Officer (CInO) of Gongos, Inc., Greg is charged with accelerating the future of everything – from trends and foresights to product innovation and development, to the company’s growth and performance. Greg thrives on exploring societal and technological shifts that point to disruptive ways to create value for consumers and resilience for organizations. Greg leads the company’s Innovation Think Tank – a cross-generational team that fosters a culture of innovation and guides long-term strategy in shaping the decision intelligence space. A former research practitioner with over 20 years of experience under his belt, Greg is a visionary at heart. He believes our industry is in the midst of a revolution, and plans to help pave the way. He holds an M.A. in Humanistic and Clinical Psychology from the Michigan School of Professional Psychology, and a B.S. in Industrial Administration, Marketing and Finance Concentrations from Kettering University.

Google’s John Mueller was asked if using stock photography on your pages would hurt the page’s chances of ranking well in Google search. Do you need a unique image per page? John Mueller responded on Twitter that “it doesn’t matter for web-search directly.”

He did add “for image search, if it’s the same image as used in many places, it’ll be harder.”

Here are those tweets:

It doesn’t matter for web-search directly. For image search, if it’s the same image as used in many places, it’ll be harder. (There’s also the potential impact on users, after search happens, eg: does it affect conversions if your team photo is obvious stock photography?)

— John (@JohnMu) June 27, 2020

Now, John did say in December 2018 that it is worthwhile to have unique images on your pages. But again, here he was referencing image search ranking. He said:

I think having a unique photo is definitely a good idea. Because if it’s the same photo that’s reused across a number of different articles, we’ll we’ll pick one of those articles for for image search to show as the landing page for that. So you’re kind of in the same group with everyone else if it’s the same photo. Whereas if it’s a different photo, then we we can show it separately in image search but that’s specific to image search.

It’s not the case that if you have good images that they will make your site rank better in web search. So it’s kind of kind of separate there. But that’s something where but sometimes good images show up as well in the normal search results. With like or you have the the images one bar on top or something like that. So I I think if you have a chance to to have your own images I think that’s definitely worthwhile.

I was trying to find that classic stock photo everyone uses with a smiling call center representative, but I used a cat instead for this story.

Forum discussion at Twitter.

Social media is a fun place where we can engage with our audience on a daily basis. You’re probably already familiar with the benefits of social media, which means you maintain consistent social media activity. 

Unfortunately, during troubling times like the COVID-19 outbreak that we’re experiencing today, businesses often struggle to maintain an active social media presence. Your budget is low, the workforce is limited, and there’s usually little motivation to produce content with everything that’s going on around you. 

Still, it’s not that hard to maintain social media activity during a crisis, and it is perhaps the only sensible thing we can do. It’s important to stay engaged with our audience during the crisis, and there’s a lot we can do to accomplish that. 

Here are four strategies to maintain your social media activity during a crisis.

1. Repurposing content

If we are unable to create fresh content, we can always work with what we already have. If you had a well-built content strategy prior to the crisis, then chances are you have a lot of pieces to work with. Our goal here is to repurpose existing content into something fresh. 

Start with what you already have: a podcast, a video log, a long-form blog article, a sales letter, anything works. Try to collect all long-form, pillar content that you have. Next, we’re going to use and repurpose that content to create fresh content. A vlog turns into a blog, a blog into an email, an email into a tweet, and so on… you get the point.

Gary Vaynerchuk is a master of repurposing content, he also popularized the content pyramid model that is based on this idea. Gary says he can create 30 fresh pieces of content to be used across his channels just from a single daily episode of his show.  

Using a single piece of content, you can create fresh content for your social media accounts, and it doesn’t have to be a repost. You can repurpose a piece of content to tweet some bits on Twitter, start a discussion on Facebook, post an edited clip on Instagram, or share a concise blog post on LinkedIn. And boom, there’s your content.

Also, there’s no need to feel like a fraud for repurposing ‘used’ content. Most of your followers won’t remember your older posts, and they could always use a reminder, especially during a crisis. Even if we have nothing ‘new’ to say, we can still share our insights from the past. To give your old content a fresh look, you can add some eye-catching visuals to it. You can take the help of a web designing firm to create visuals that can get noticed in crowded social media feeds. 

2. Make use of content creation tools

With everything slowing down, it’s hard to create enough content all on your own. In the past few years, we saw a lot of content creation tools and templates come to life, and perhaps it’s time to make good use of them. Content creations tools help us to minimize the time, budget, and effort needed to create content, and now we need them more than ever.

Depending on your needs, there are various tools to choose from:

• For research, you can make use of Google Drive’s Research Tool to conduct quick research, all it takes is clicking a simple ‘Explore’ button in the bottom right. Also, ‘Site: search’ function is another useful tool accessible from the browser.
• If you need help writing posts for your social media account, you can use writing tools like Evernote to take notes, Grammarly to catch errors,  WriteRack to tweetstorm. 
• If you want to post visual media then you have to try out tools like PicsArt. These tools are easy to use, and you can create great visual content in less than five minutes. Instead of spending hours on design, all you have to do is choose a template and fill it with your brand graphics.

It takes a lot of effort to create great social media content, but we can always make use of content creation tools to save some time or get a few creative ideas. 

3. Utilize user-generated content

User-generated content (UGC) is content created by people rather than brands, which means you don’t have to create anything. Utilizing UGC is incredibly important for social media, and it can be used to fill the gaps in your content strategy. Brands may not be able to create their own content during the crisis, but can always rely on user-generated content. 

The type of content you repost will vary depending on the media. 

Instagram: The king of user-generated content, Instagram has all kinds of options for brands to share content created by users. You can repost to your own profile, share images on your story, and easily browse using #hashtags and the Explore function. Aerie is a great example of how this should work:

Facebook: Facebook is a fantastic network for sharing stories and videos with your audience. You can invite your fans to contribute stories, images, or videos and use it to invite discussion and engage with the rest of your audience.

Twitter: A great place to utilize user-generated content, Twitter makes it easy with #hashtags and the “Retweet” function. You can simply retweet users and add your own comments to spark a discussion. Food brands do a great job on Twitter:

when you don’t get the toy you wanted in the kids meal https://t.co/dJc4yiAoB1

— Burger King (@BurgerKing) October 8, 2019

LinkedIn: Professionals love LinkedIn, and you can use LinkedIn to promote user content that’s relevant to your brand. You can repost the content or feature some users in your blog posts. 

If you want to search for location-specific content, you can always use a VPN service to gain access to content specific to a certain location. This method helps you to understand how your audience sees things, and you can tailor your content to meet their personal needs. 

4. Keep up with the updates

Posting relevant content is important, but don’t forget to post personal updates about your business. Your audience may want to know how you’re doing, whether there will be disruptions in service, and what to expect in the coming days. 

To add on to that, make sure you understand your position during a crisis. If you’re in the middle of it, you can provide daily updates on how your local community is dealing with the crisis, and that’s a good way to build a relationship with your audience. 

Lastly, don’t forget to show compassion for the victims, and you can even use one of the content tools to create supportive posts and remind your audience that you’re thinking of them. 

What’s your take?

What do you think about the ongoing crisis and what is your strategy to maintain your social media activity in the upcoming weeks?

Ah, the age-old debate.

• Should I bid on my brand terms or not?
• Why do I have to pay to get the traffic I earned?!?!
• Why are Google and Bing holding my brand hostage?

These, along with a myriad of other questions around brand bidding, seem to pop up every few years.

Rising CPCs, an evolving SERP, and a few lawsuits have led us to a new breakpoint.

I’m going to do my best to answer the most common questions I’ve encountered over my career.

This should help guide you on how (and if) you should bid on your branded terms.

(For the sake of consistency, I’m going to pretend my brand is Seinfeld throughout this post. Many references to follow.)

1. Should You Bid on Brand Terms in PPC?

Yes. And also a little bit no.

But mostly yes.

But not always, and not everything.

It’s not as binary a decision as it’s made out to be, nor should we treat all branded terms equal.

We must view brand terms the same as any other terms in SEM.

They need to have a goal, KPIs, and a strategy to achieve success.

Don’t bid terms to the ceiling and issue a blank check because you want a 100% impression share.

There are smarter (and more profitable) ways to operate.

Ask yourself: would I make this decision if the query was “best tv show ever” instead of “Seinfeld”?

If the answer is no… well, you get the point.

2. How Should I Structure My Branded PPC Campaigns?

Structure your campaigns… well, the same way you structure everything else: tiered in order of volume and budget priority.

My recommendations will (likely) be against engine recommendations, but so be it.

I’m a grownup and I do what I want.

Generally speaking, you’d want to have 4-5 “sets” of branded terms.

It’s up to you whether you structure them into campaigns or ad groups, but I’d recommend breaking them out either way.

Isolate Core + Navigational Terms Should Into Their Own Areas

By navigational, I’m talking about terms like “www Seinfeld com” or “Seinfeld com”.

When I say core, I’m speaking about things like “Seinfeld show”, “Seinfeld”, “Jerry Seinfeld show”, and so on.

These are the highest volume, the highest propensity to convert and likely lower incrementality.

Promotional or Discount Queries

These should be addressed but not necessarily by you… more to come on this.

Brand Adjacent/Sub Brand

This includes names of products or terms with a branded intent.

I’m thinking things like “Kramer’s lobster shirt”, “Jiffy Park t-shirt”, or “Urban Sombrero”*.

These aren’t your brand in the conventional sense. But, if a consumer is searching for a specific item by name, they should fall in this bucket.post

*For those who don’t know, these are Seinfeld apparel items worn by characters.

Branded Research

This covers queries like “Seinfeld reviews” or “Seinfeld vs. Friends”.

These are closer to mid or bottom funnel and may not represent a person who’s ready to buy right this second.

I could see arguments we shouldn’t consider them branded terms at all. That said, while these may not convert as well, they’re likely to have a high level of incrementality.

3. Should I Be in the Absolute Top Position, 100% of the Time, for Every Brand Mention?

No. I outlined this in-depth some time ago in a post around forecasting.

Every click (branded included) is more expensive than the last.

As you get closer and closer to 100% impression share, those clicks will get exponentially more expensive

The structure above presents a sound framework to maximize efficient volume.

If you need a target (big blanket statement warning) generally somewhere in the low 90-95% mark is where returns start to diminish.

That’s not saying you can’t make a profit above that level, but you’ll likely see ROI/ROAS start to diminish.

But again, test.

4. Should I Automate Bids on Brand Campaigns?

Yes, but test which will work best for you.

Most forms of smart bidding have their place depending on your goals.

Target Impression Share has proven fruitful with an aim towards that 90-95% sweet spot.

It keeps CPCs at a reasonable level and handles most of the competitive variables well.

By the same token, tCPA or tROAS have been beneficial in the right scenarios depending on our clients’ goals.

Make sure to set caps to prevent your CPCs from spiking too much.

This may be the only scenario where I still endorse (somewhat) manual bidding as well.

Variance for brand campaigns is lower than other broader terms so your bids likely won’t need to change as much.

Basically, it’s up to you.

5. Should I Use Enhanced CPC (eCPC) for Brand Keywords?

Test it, but I don’t recommend it as a blanket “must do.”

Think of it this way; eCPC aims to increase conversions by bidding more when a conversion is more likely and vice versa.

Say you already have 97-99% top impression share for your best terms.

eCPC will (almost) always assume a conversion is likely to happen and will (almost) always bid up.

You then pay more for the same result.

I have seen it work well in competitive industries where impression share is, shall we say, harder to come by.

But, test it, don’t assume.

6. If I Turn off Branded PPC Terms, Wouldn’t I Get All the Traffic & Conversions Anyway?

“Well these people were looking for me anyway, so they’ll find me no matter what.”

No.

This is the most common argument against bidding on brand terms.

It’s one of the more challenging things to test, and the most likely to prove a case for bidding.

Incrementality is the proportion of outcomes that would happen without intervention. That is, how many conversions would we get if we did nothing.

Incrementality testing has varied in every one of the dozens of brand turnoff tests I’ve run in my career.

I’ve seen incrementality as low as 15-20% (e.g., 80-85% of conversions would’ve happened without ads) with a naked SERP.

Though, the frequency of a truly naked SERP is next to zero nowadays.

I suspect this incrementality figure is higher in modern times.

I’ve also seen it as high as 85% for competitive industries like mattresses, fitness, and home services.

I’ve outlined a few examples below, showcasing how we ran the tests and the result.

We Tested for a Supplement Company with a Short Purchase Path

Automated rules turned off branded ads every other day for 60 days.

The total return on search investment was our success metric: (organic + paid revenue)/search spend.

Days with brand ads on yielded 28% more total search conversions.

Factoring in the margin and the cost of ads, brand investment was ROI positive, albeit closer to 2:1 vs. the 10:1 reported in the engine

A National Home Improvement Company Left the Decision to Invest or Not During COVID in the Hands of Franchisees

• Approximately half turned off branded ads while half remained on.
• Franchises that paused brand search performed 50% worse than those who kept it turned on across all engines.
• The team conducted a regression analysis to figure out what happened to organic ads during the same time. Those who kept ads on saw more organic leads than those who turned off (R2 = 0.45), indicating there was a positive impact while ads were on.
  • The direct result was ROI positive – the halo effect made the decision a no-brainer for franchisees.

A Well-Known Marketplace Wanted to See If They Could Save a Few Dollars by Removing Brand Terms

• Their CPCs were low (in the five-cent range) and a naked SERP outside of shopping ads. We ran a geo-holdout test for a period of six weeks. We excluded ~100 ZIP codes from branded ads and compared against the performance of 100 like ZIP codes. The findings were dramatic:
  • Approximately 44% of branded clicks were incremental, vanishing without ads running
  • CPCs on incremental clicks were 7x higher than the cost of non-incremental clicks. It’s worth noting in this case non-incremental clicks cost under a penny each.
    • ROAS on branded terms in the engine UI was around 80:1 (which is insane) – iROAS, that is the incremental revenue/total spend) was around 12:1.

I could go on, as there are dozens of examples around.

TL;DR: The incrementality of branded terms varies but is (in my experience) always positive.

7. Should I Update Ads for Every Promotion?

I wouldn’t recommend it.

There are two main challenges, one psychological and one PPC-ical:

• Search engines favor ad history. When you throw a new ad for every 10% off sale, you’re resetting history and making the engine guess what will happen. Generally, there’s a CPC “blip” with each new ad until the engine figures out what to do with it. If you have a short-term sale, that blip may last the whole sale and reduce efficiency.
• Consumer intent from a branded query is different from a sale ad vs. a non-sale ad. First-time customers from a promotional ad likely have lower incrementality and a lower LTV. They’re more likely to become “sale only” customers who only buy cheap.

You’re a great PPC-er so you’ve been testing for years – odds are your evergreen branded ads are quite strong.

Unless your sale has a dramatic effect on performance, stick to promotion extensions or sitelinks.

If you must run a promotional ad, keep your existing branded ad live in tandem.

8. Should I Let Resellers & Affiliates Bid on My Brand Terms?

Controversial opinion time!

Yes, with restrictions.

Allowing a few selected partners on your branded terms accomplishes three things:

• It builds a moat for competitors. That’s not saying competitors can’t or won’t usurp your branded terms, but there are more layers of defense.
• They present options that are more likely to align with consumer intent. I have a sneaking suspicion that someone searching for “Seinfeld Coupons” already intended on buying. Let an affiliate take that click – it’ll likely be a higher CPC that you wouldn’t make much money on. Plus, it’s a better reflection of consumer intent and a better experience.
• When done well, they won’t raise your CPCs. You hold the cards in the relationship with affiliates. You can set the terms of their presence to ensure CPCs don’t spike.

So Should I Bid on Branded Terms?

Yes.

But as with all things SEM, you cannot evaluate your success and ROI in a vacuum.

Instead, look at the net impact on your business (including the halo effect) to make sure your branded bidding strategy aligns with goals.

More Resources:

Ethiopia police were on Wednesday detaining leading opposition politician Jawar Mohammed, a move that risks inflaming ethnic tensions that have led to multiple deaths during protests in the capital and surrounds.

The capital Addis Ababa was rocked by a second day of protests which erupted on Tuesday following the killing of Hachalu Hundessa, a popular Oromo singer who was gunned down Monday night.

At least eight people have been killed, according to an AFP tally, in the Oromia region which surrounds Addis Ababa and is the heartland of the country’s largest ethnic group, the Oromo.

Jawar, a former media mogul who recently joined the opposition Oromo Federalist Congress, was arrested in Addis Ababa along with 34 other people, federal police commissioner Endeshaw Tassew said in a statement late Tuesday.

Endeshaw said that as Hachalu’s body was being transported to his native town of Ambo for burial, Jawar and his supporters intercepted it and tried to return it to Addis Ababa, where a clash ensued.

“There was a disturbance between federal security forces and others, and in the process one member of the Oromia special police force was killed,” Endeshaw said.

“The security forces have taken eight Kalashnikovs, five pistols and nine radio transmitters from Jawar Mohammed’s car,” he said of the arrest.

The Oromo Media Network — which was founded by Jawar before he left to become a politician — reported there was a call for mass protests until he and the others were released.

– ‘A dangerous situation’ –

The internet remained cut off for a second day in a government bid to curb the unrest.

In October, reports that the government was attempting to remove Jawar’s security detail kicked off days of violence that left more than 80 people dead.

Hachalu’s music gave voice to Oromo feelings of marginalisation that were at the core of years of anti-government protests that swept Prime Minister Abiy Ahmed to power in 2018.

Map of Ethiopia and its regions

Simon MALFATTO, AFP

The motive for his killing has not been identified, but police have said “some suspects” have been arrested.

In Addis Ababa — where protesters grouped at several points around the city — security forces on Wednesday fired into the air to disperse demonstrators who were approaching a statue of Emperor Menelik II, widely seen as the creator of modern-day Ethiopia.

Oromo nationalists see Menelik as a driving force behind their perceived marginalisation, and Hachalu called earlier this month for the statue to be pulled down.

On Tuesday protesters in Harar in eastern Ethiopia pulled down a statue of Ras Mekonnen, the father of Emperor Haile Selassie, a doctor in the city told AFP on condition of anonymity, also reporting the death of one person during protests.

– ‘Everything is closed’ –

In the town of Nekemte in western Ethiopia, a doctor at the Wollega University Hospital, Negeo Tesfye, told AFP: “Yesterday there was a clash between protesters and local police, three people were then shot by regional special forces. Two of those people died.”

“Currently everything is closed down, there is no transportation, people are not moving around,” he said.

Medical sources and relatives on Tuesday reported three deaths in central Adama, and another in Western Hararge.

Oromo living in the United States have gathered in Minnesota to protest the death of musician and activist Hachalu Hundessa

Stephen Maturen, GETTY IMAGES NORTH AMERICA/AFP

Federal police said several people had also been killed during three grenade attacks in the capital, without giving exact figures.

Ethiopia, an ethnic melting pot of 100 million people, has battled deadly intercommunal tensions in recent years, a major threat to efforts by Prime Minister Abiy Ahmed — himself an Oromo — to bring about democratic reforms in a country long ruled with an iron fist from Addis Ababa.

“The assassination of an important Oromo musician, subsequent protests which have in places involved property destruction and security forces using lethal force, and the arrest of Oromo leaders, creates a dangerous situation and is another blow to Ethiopia’s troubled transition,” said William Davison, an analyst with the International Crisis Group.

The lack of diversity in the advertising and marketing industry is not a new issue. Despite loads of research on the business benefits of diverse teams, there’s been talk but little action for years to increase Black and minority representation in the industry.

During this session of Live with Search Engine Land, which took place during SMX Next last week, I discussed how to accomplish change in your organization with:

• CJ Bland, co-­founder, CEO and principal consultant of the Minority Professional Network, which specializes in DE&I recruiting and retention, marketing, training/speaking and consulting.
• Zenia Johnson, an account lead at digital agency 3Q Digital who specializes in social media and is passionate about fostering inclusivity and diversity in the digital technology space.
• Jackie Leung, director of talent acquisition at digital agency Wpromote. She oversees all hiring practices nationwide and leads the agency’s Diversity Equity and Inclusion Initiative.

We discussed why diversity is good business, what intentional commitment to diversity, equity and inclusion (DE&I) initiatives looks like, recruiting, hiring and retainment practices to achieve diversity goals and how employees can initiate change in their organizations. Watch the full session above.

For more on this topic, see the accompanying article: Actionable ways to drive diversity, equity and inclusion in your marketing organizations.

Live with Search Engine Land’s weekly meetups are about giving great marketers a platform to inform, support and convene our global community. If you have an idea for a session or would like to join a panel, email kbushman@thirddoormedia.com.

This story first appeared on Search Engine Land.

About The Author

Ginny Marvin is Third Door Media’s Editor-in-Chief, running the day to day editorial operations across all publications and overseeing paid media coverage. Ginny Marvin writes about paid digital advertising and analytics news and trends for Search Engine Land, Marketing Land and MarTech Today. With more than 15 years of marketing experience, Ginny has held both in-house and agency management positions. She can be found on Twitter as @ginnymarvin.

Yesterday Google has removed the “Sponsored” label on the Google Shopping Ads and replaced it with the black “Ads” label you see on normal Google Search Ads. Ginny Marvin from Search Engine Land said this is “streamlining” the ads label.

Here are before and after screen shots that I saw myself on the desktop results.

Old “Sponsored” Google Ads Label (click to enlarge):

New “Ads” Google Ads Label (click to enlarge):

Ginny also produced an excellent history of Google Ads labels.

I prefer it read “Ads” versus “Sponsored” – so I am happy with this change.

Forum discussion at Twitter.

Please forgive the fact that I’m tweaking a tired adage, but the message is true: Building your brand authority doesn’t happen overnight.

I was reminded of this fact very recently while scrolling through LinkedIn: 

Ongoing marketing efforts are needed to tell an authoritative story and build trust in potential customers. It can always make a difference when someone is deciding between two companies, and it’s even more important with B2B, since those products/services tend to involve a higher cost.

Here’s how you can go about utilizing digital marketing to increase your brand authority.

Note: I’m going to focus on the content itself, but earning backlinks — which is significantly easier to do with high-quality content — is a primary way to indicate to Google that other sites trust you, which signals that you’re more authoritative. Prioritizing your backlink portfolio will dramatically help you in all other authority-building efforts.

Gauge your brand authority level

Don’t assume you already understand how you’re viewed by your audience. Instead, before launching into any marketing strategies, check the data to get a sense of how you’re being perceived.

• Have your branded searches increased or decreased? What search terms are people pairing with your brand? 
• How are your customers or leads finding out about your brand? Was it from authoritative interviews or content you put out there or some other way?
• Are you ever mentioned in the media? If you haven’t already, set up Google Alerts for your brand name and any prominent, public-facing employees. 

Another interesting consideration is: Who are the current authorities in your space? Are you aware of them all?

The first way to identify this is to type into Google the phrases you wish you ranked for and see who is ranking for those terms. Sometimes it’s the competitors you knew about, but sometimes other sites have climbed up the authority ladder.

Additionally, you can use tools like SparkToro to search your topic area and see where your audience is going for information. 

If you search for your vertical, you can then see the most popular publications, podcasts, social channels, and more visited by the audience interested in your vertical.

Then the question becomes, are you on these lists? If not, who is and why? What are they doing well? You can aim to be featured on these different media outlets, as you know they appeal to your target audience.

Identify your audience’s questions

If you answer your audience’s questions, they’ll start to trust you and see you as an authority.

The concept sounds simple, and it is. But the execution is harder. First, how do you find out what their questions are?

Here are a few ways:

• Tools like Answer the Public and BuzzSumo’s Questions will show you what people are asking based on different keywords you enter. 

Snapshot of BuzzSumo’s Questions tool

• Keyword research can reveal the types of challenges people are facing. Don’t just look at keyword volume — look at “People also ask”. Get lost for a little while, clicking on various questions and related keywords. (Keywords Everywhere is a cool tool for search volume/competition, as is Keyword Surfer).
• Talk to your sales team about what common questions are coming up. Have you answered these with content? Do they speak to the greater problems your audience faces?
• Brush up on your audience personas. Different segments of your audience may have different problems. See if you’ve been accidentally neglecting a segment.

Once you have a solid list of the questions your audience has, you can work on effectively answering those questions.

Answer the questions with authoritative content

Once you know what you want to write about, how do you make it authoritative?

First of all, your methodology matters. Do your own original research whenever possible. Content backed by data is inherently more trustworthy than content based on opinion. If you’re featuring opinion, make sure it’s someone who can prove their expertise through their past experience.

Secondly, the content has to be created in a way that conveys authority:

• It shouldn’t have any grammar or spelling errors
• If it’s time-sensitive in any way, it needs a date on the article so people know exactly when it was written and thus the content can be put in its proper context
• Sources should all be cited
• The design should be clean and easy to read
• The structure and navigation should be well-thought-out and provide insight into exactly what readers will learn
• All information should be backed up with explanations and facts
• If your piece was written by experts, provide their name and bio

Let’s take a look at some examples. I pulled the top organic text and video results for the query “how to choose a bike.” (I’m thinking about buying a bike, so I’m finding myself using a lot of bike-related examples as of late…)

REI’s article, “How to Choose a Bike,” ranks number one. I use REI examples a lot because I think they have a fantastic content strategy by using their expertise to answer all kinds of questions their customers could have.

But let’s focus specifically on what makes this article seem authoritative. 

First, it’s well organized and clearly outlined, even including a table where you can get the top-level information very quickly. Having a well-thought-out structure and design is a visual indication of knowledge and understanding of a topic.

They also have a section at the bottom labeled “Contributing Experts” so you know exactly who put the guide together and what experience they have.

Finally, they responded to all of their comments, providing additional information to the people who had further questions.

Now let’s check out the top video result, which is from 2013, meaning people have found it useful for more than six years. What about it feels authoritative?

For one, look at how he outlines right at the beginning what the video will cover, setting proper expectations and indicating a solid knowledge of the subject.

Additionally, he doesn’t just list the features of the different types; he explains the usefulness of those features to help you make a more informed decision.

There are a few other techniques to display authority, as well. Andy Crestodina recommends including quotes and tips from other thought leaders in your piece. You can also get third-party validation for your content in the form of testimonials, reviews, or asking influencers to share what you created. The point here is to showcase that you associate with experts and that other people trust you.

Conclusion

It’ll take time and effort, but once you’re an authority, every other aspect of your marketing will gain more traction. Consider how to build authority into all of your digital marketing, and you’ll have the potential to amplify your results even further.

Amanda Milligan is the Marketing Director at Fractl, a prominent growth marketing agency that’s worked with Fortune 500 companies and boutique businesses.

Porsche Design will let consumers customize their watches the way they customize Porsche cars, from trim to color and everything in between. Image courtesy of Porsche Design

E-A-T has been a hot topic when it comes to on-site content.

But why, how, and when should you consider concepts in E-A-T with link building efforts?

Your site cannot have E-A-T without a link and brand mention profile.

In the Search Quality Rater Guidelines, Google states that “sources of reputation” are “news articles, Wikipedia, articles, blog posts, magazine articles, forum discussions, and ratings” when establishing Page Rating.

And it’s easy to determine that E-A-T can, in certain niches and to varying degrees, be leveraged in evaluating a site’s fit to secure a link.

But, Google’s Ben Gomes said in a 2018 interview with CNBC that:

“You can view the rater guidelines as where we want the search algorithm to go.”

“They don’t tell you how the algorithm is ranking results, but they fundamentally show what the algorithm should do.”

Even though E-A-T doesn’t directly have an impact on ranking, it can result in creating signals that drive direct ranking improvement.

Still, it’s important to use your own experience when understanding what works and doesn’t work in your niche.

If you follow Google’s recommendations or guidelines strictly, you’ll always be playing the long-game or even lack any results, when it comes to link building.

And it’s clear that Google cannot even implement an algorithmic interpretation that has the strictest interpretations of these guidelines.

A great example of this is under recipes.

My eight-year-old daughter and I were looking for banana bread recipes in Google.

We found the one in the screenshot below from Allrecipes.

I found that this recipe is actually way off on the bake when we baked it, by more than 30 minutes. But it had more than 15K reviews.

I can dig up a hundred similar examples fairly easily.

So there is a major caveat when applying my interpretation of E-A-T to guide link acquisition.

Don’t use E-A-T guidelines to strictly evaluate prospective sites in every niche.

When to Use E-A-T Standards for Link Prospects?

There are three major considerations before even bothering to review a site against your guidelines, under a fair interpretation of Google’s search quality raters guidelines.

• If the sites are completely useless, then no need to do an in-depth analysis.
• Not every niche or scenario requires a strict analysis of a site to build a link.
• A natural link profile doesn’t just have E-A-T sites.

Let’s start with a “useless site,” as there is no point in further review of a site if it is not designed to create value but is “made for SEO.”

Google’s page quality guidelines state that:

“Websites or pages without any beneficial purpose, including pages that are created with no attempt to help users, or pages that potentially spread hate, cause harm, or misinform or deceive users, should receive the Lowest rating. No further assessment is necessary.”

I do not believe in having a strict interpretation of the phrase “beneficial purpose.”

My interpretation is that if the site has no relevant content that helps users then it’s useless.

But it isn’t so clear.

For example, The Million Dollar Homepage had no value but to sell ads so a guy could make a million dollars by selling ads.

This site has over 94,000 backlinks (Ahrefs) but only ranks for 1,200 keywords.

This is an extreme example of a site I’d just avoid.

As for point 2, there are certain niches that require much stricter review than others.

YMYL sites have much stricter standards than other niches.

In the guidelines, Google provides some guidance on what constitutes a completely useless site.

In the guidelines, Google explains that YMYL vs other sources will have different sources of reputation.

The last point is just a reality check for you people that review your link profile, or even build disavow lists with a strict interpretation of what is “high” or “low” quality.

If you’ve analyzed backlink profiles for competitive niches, then you’ve certainly seen a mix of links that you may consider verily degrees of quality and with a wide metric variation.

The screenshot from Ahrefs shows that bestcolleges.com’s page that ranks for “best online colleges for psychology” has a mix of link quality and domain rating (DR).

There are certainly many lurking and confounding variables that also have an impact, but it’s so commonplace that you cannot ignore this fact.

What Are the Criteria for Evaluation?

My team and I have built a score with over 50 variables we delivered called PureGrade, to understand the likelihood of the site having a positive, negative, or neutral impact based on main content (MC) and supplemental content.

That said, Google uses an interesting scale that, in my opinion, provides way too much freedom to raters.

My team has used these criteria to rank sites that don’t necessarily qualify for YMYL content.

This has been built based on a manual review of over 60,000 sites and thousands of pages, across hundreds of enterprise-level projects.

And, Moz’s spam score has over “27 common features” that they claim correlates with a significant amount of banned or penalized sites.

I don’t recommend using Moz’s spam score, PureGrade, or other single metrics as gospel for disavowing or prospecting new sites.

However, it helps to illustrate the point that we need to evaluate sites against a set of criteria and not only one or two.

Google guides its raters to use a scale of lowest, lowest+, high+, high, medium+, medium, low+, lowest +, and lowest.

It’s really important to note that this scale is used to rate the ranking site and not the linking site. This doesn’t mean you can’t use this scale to evaluate prospects.

For the purposes of this article, which is to make a connection between the quality rater guidelines and link-prospecting review criteria, I’ll only be listing out elements of a site based on E-A-T from ideally what Google’s, or search engines, “algorithm should do,” and not what I’ve seen work to drive ranking improvements.

The rater guidelines only press on the idea of finding an expert in the niche and verifying that they are experts.

These core elements to review are:

• About us, contact, or customer service info.
• Outbound link profile.
• A “positive reputation” of the main or supplemental content.

Beyond just finding the authorship, it’s important to evaluate the MC pages, and one key indicator is whether the site mentions “buy dofollow links” anywhere on the site.

If the site does say this, then it’s likely that most of the articles are not from reputable sources and you should avoid it.

This is not the case when the site mentions “paid guest posting.” This can mean they offer sponsored posts or affiliate links.

This is just how sites make money, and can be separate from the main content.

With an outbound link profile, search for overly commercial links.

These are typically links that surprise the user when clicked by funneling the user to the homepage or transactional pages.

Finally, the “positive reputation,” is much harder to establish.

Use a mix of metrics from Ahrefs, Majestic, and Moz to understand the site reputation profile.

But you can also use tools like BuzzSumo to identify if the articles of the prospect get shared or have engagement.

A Final Note

It’s really important to note from the rater guidelines that:

“Frequently, you will find little or no information about the reputation of a website for a small organization. This is not indicative of positive or negative reputation. Many small, local businesses or community organizations have a small “web presence” and rely on word of mouth, not online reviews. For these smaller businesses and organizations, lack of reputation should not be considered an indication of low page quality.”

When evaluating micro-influencers or small blogs for prospects, you have to realize it’s not always critical to be strict.

More Resources:

Image Credits

In-Post Image: Created by author, June 2020
All screenshots taken by author, June 2020