The search landscape is always changing. Anyone who has been in SEO for even a few years will know this to be true.
One of the most significant changes that we are likely to see in 2019 is a continued move to more people using voice search. If we look back to 2016 Google said that around 20% of mobile queries at that time were voice searches.
Google hasn’t released any further data since that time but with their investment in Google Home and recent algorithm updates that look at NLP it would suggest that this is still on an upward trend.
In a study from canalys from 2018, we can see that Google shipped around 3.2 million of their Google Home and Google Home Mini devices in Q1 of 2018 – and that is set to rise.
It’s easy to see why Google’s market share has increased in this area. This study from Stone Temple took 5000 questions and asked them to Alexa, Cortana Invoke, Google Assistant on Google Home, Google Assistant on Smartphone and Siri.
The Google products came out top in the test.
They also made fewer mistakes than the competition.
The investment that Google is making in this area – coupled with the ongoing growth of smartphone ownership – show that SEO for Voice Search is something you cannot ignore.
You need to be optimizing for Voice Search now so that you don’t get left behind.
Why You Need To Make Optimizing For Voice Search An SEO Priority
Most SEO’s will have heard of the Search Quality Rater Guidelines, the ridiculously long document that helps Google manually check the quality of their search results.
But did you know that Google had Voice Search Rater Guidelines? Google is taking this very seriously.
We also know that featured snippets, and related questions are interlinked with Voice Search (more on this later) and at the time of writing Mozcast reported that 82.6% of the search results they analyze had related questions and 14% had featured snippets.
If we look at data from SEMrush, we can see that 47.7% of mobile search results in the US have People Also Ask features.
We can also see that 7.17% of mobile search results in the US have featured snippets.
The data is always going to differ depending on the tool you use, but it’s obvious that these features are here to stay. As an SEO you need to be thinking about how you are going to capitalize on this. Don’t think of it as Google stealing organic traffic – think of it as an opportunity to make sure your SEO is future proof.
Keyword Research For Voice Search
We know that a lot of people that are using Voice Search are asking questions. Questions that they need quick answers to. So how do we find the questions that people are asking of Google? How do you do your keyword research?
I have a couple of favorite tools that I use for this. Let me walk you through how you can use them for optimizing for Voice Search.
So how does it work?
When you arrive at the site, you are pestered by a gentleman they like to call The Seeker. To make sure he doesn’t get too impatient you will need to drop a seed keyword into the search box – you need to be thinking high level here, pick keywords that define your market for the best results.
Within a few seconds, you will be presented with a wealth of data that you can mine for inspiration. These ideas are split into 5 categories.
When doing keyword research for Voice Search, I tend to concentrate on the questions section, although it is always worth looking through the other categories for inspiration.
The visualization is a great place to start as it draws you in, but I highly recommend downloading the CSV so that you can add additional metrics to your research.
From here you can take that research and upload it into your favorite keyword research tool. I like to use Keyword Keg for this. It has a simple upload function and delivers the data quickly.
While a lot of the metrics are desktop based, this gives you a foundation to build on. You can sort your data at this point rather than having a list of questions and not knowing where to start.
The metrics it features are unparalleled for the price of the product (hint: it’s very cheap) and covers.
- Traffic Value
- SEO Difficulty
- CTR Scope
- Keyword Power
- Words and Length of Query
- OnPage Difficulty
- OffPage Difficulty
If you have a long list of questions it can also help you filter out irrelevant keywords with a negative filter.
Exporting all of this data gives you something much more actionable than a list of questions.
Made by the same people as Keyword Keg, Keywords Everywhere is a Chrome extension that can help you come up with even more potential ideas.
The image below – for the question “how will voice search affect SEO?” – brings Related Keywords and People Also Search For suggestions into the search results.
Many of these can be used to tailor your keyword research further. The tool has an export to CSV function so that you can add them to Keyword Keg with ease.
Combining this with the questions in the main People Also Ask box, and you are gathering real data to fuel your Voice Search SEO and question based content campaign.
Bonus tip: don’t forget that the People Also Ask box has been infinite for some time. Keep clicking for more ideas. Not all of them will be relevant to you, but there are some hidden gems that other tools just cannot surface.
SEO Tactics To Improve Voice Search SEO
In his study of 10000 Google Home search results, Backlinko investigated the potential impact of 11 factors that could lead to Voice Search ranking.
Brian Dean specifically looked at:
- PageSpeed – the average Voice Search page loads in 4.6 seconds
- HTTPS – 70.4% of the results were HTTPS
- The typical word length of answers – Brian found that the average length of the answer was 29 words
- Schema – 36.4% of the results came from pages that used Schema – only slightly than higher than the average of 31.3%
- Domain Authority – the average Ahrefs Domain Rating was 76.8
- Social influence – the average page that triggered a Google Home result had been shared on Facebook 1199 times and 44 times on Twitter
- Ease of content to read – the average reading level discovered on the pages was 9th Grade
- Title Tag correlation – most of the pages didn’t feature the exact question in their Title Tag
- Length of content that Google has sourced the answer from – the average length was 2312 words
- Desktop ranking correlation – Brian found that there was a strong correlation between top 3 rankings on desktop and getting a Google Home result
- Featured snippets – 40.7% of the results also had the featured snippet in the SERPs
In an ideal world, we would focus on as many of the findings as we could, but I know that isn’t always possible. It can be tough to get client or stakeholder buy-in. For example, drastically improving PageSpeed on sites is often down to how much development time you have within your team or budget.
So what do I think you should concentrate on?
What are the tactics that you can action today?
For me, it comes down to two main areas. Focus on trying to win featured snippets and improving the content on your pages.
You now have the keyword research done and dusted so let’s see what you need to do when you have decided on what content to create and the questions you want to answer.
Focus On Featured Snippets
Featured Snippets have been around for some time now, first appearing all the way back in January of 2014. Even if you don’t do SEO on a regular basis, you will have come across them in the search results pages. I probably don’t need to show you, but they look like this.
If you use Voice Search to search for “what is a Title Tag?” you will hear that this Featured Snippet is the one that is used in the Google Home results as well.
The connection between desktop Featured Snippets and Voice Search is there for all to see – as Brian Dean pointed out in his study. SEOs have been optimizing for Featured Snippets for some time, and those that have been successful so far will be seeing some benefit from Voice Search as well. If you want to capitalize on the correlation between Featured Snippets and Voice Search you need to be acting now.
So how do you go about optimizing for Featured Snippets?
I have found the following tactics incredibly useful over the last few years.
- Study the search results of the questions you are trying to answer. I like to note down the number of words in the answer that already has the featured snippet and answer that at the top of my content. Even if the content is longer, you need that short, sharp answer at the top of your page.
- Study the formatting and make sure you are including that formatting in your post. If Google is showing a table for the top Featured Snippet, then you need to include one. If Google is showing a list, then you need to consider adding a list to your content. Google is choosing the format so you may as well follow their lead.
- Study the People Also Ask boxes. I mentioned this earlier in the post, but it is worth mentioning again. Keep clicking and expanding those boxes to see what people are asking about for that topic. If the questions feel like they belong in the content that you are writing (and they are different enough to the original question) then include them in your content.
- Note the length of the content that Google extracts the Featured Snippet from. You can also use a tool like SurferSEO to see what the average word count for content in the top 10 is. Use this knowledge to your advantage – these are the pages that Google deems important. Find the commonalities.
- Don’t forget the importance of traditional SEO. In a study conducted by Hubspot, it was discovered that Google can and does use <h2> Tags and <h3> Tags as list items in Featured Snippets. Traditional SEO is not dead.
Bonus tip: use SEMrush to keep an eye on featured snippets – it helps you see where you have won them and where there is the opportunity to improve.
Focus On Your Content
It has been the Google mantra for years. Create great content. The problem has always been that it’s all Google ever said. They have never really given any guidance on what “great content” is.
But that’s Google for you.
The recent Medic update and some great posts from AJ Kohn and Justin Briggs dig beneath the surface and give us some clues on what it is that Google wants to see and what they are rewarding in the search results.
So what do I recommend for improving your content for desktop rankings, mobile rankings and (by default) Voice Search results?
Embrace NLP (Natural Language Processing)
So what is Natural Language Processing?
A lazy Wikipedia search tells us it is “the application of computational techniques to the analysis and synthesis of natural language and speech.”
For me, the important part of that statement is “natural language and speech.” AJ Kohn covers why he believes the Medic update was based around this in his excellent post “Algorithm Analysis in the Age of Embeddings.” It is a little too dense to go into here, so go and have a read of it – and maybe 2 or 3 times.
To quote AJ’s own TL;DR “Recent Google algorithm updates are changes to their understanding of language. Instead of focusing on E-A-T, which are not algorithmic factors, I urge you to look at the SERPs and analyze your traffic including the syntax of the queries.”
So how do you optimize for NLP?
I came across this post by Justin Briggs several months ago in which he breaks down how to optimize for NLP. It is a mammoth 3500-word guide, and I couldn’t even begin to break it down within this article, so I suggest you take some time out, grab a cup of coffee and read it.
Brevity In Your Writing
Despite what some people tell you, good writing can be learned. It just comes down to practice and unlearning the bad habits we picked up during our education. If you think back to the Brian Dean study, he discovered that the average reading level of the of the Voice Search results was 9th grade.
Well written, simple copy works and is directly tied in to the NLP that Google has introduced into the algorithm.
When I am checking my copy, I like to use a combination of the Hemmingway App and Grammarly. Both have free versions and are incredibly easy to use. In both cases, it is a simple case of adding your content to the app and following the recommendations that they have.
Here are screenshots from both tools as I use them to edit this article that you are reading now.
Tools like this can help you create more concise content. Content that is also going to rewarded by Google’s algorithm. Content that can get you Featured Snippets. Content that is likely to be used in Voice Search.
And that is what Google wants, now more than ever. We now have an idea of where Google is heading. We want to rank on mobile, we still need to rank on desktop, and we definitely want to be part of the future and also rank in Voice Search.
Questions are going to be key moving forward and have to form part of your content and SEO campaigns. You now have my blueprint for how I optimize for them and how I have achieved success with Voice Search.
I hope you do too.