What is Voice Search SEO?
Voice search is simply to use your voice to perform actions on the web. Most people were first introduced to searching by voice on Apple’s personal assistant app, Siri. Much has changed since then. Siri is still a major player in voice enabled search but there are now several more applications and devices utilizing voice search. In addition to Siri (previously powered by Bing, now powered by Google) there’s Google Assistance, Microsoft’s Cortana (powered by Bing of course) and Amazon’s Alexa (powered by Bing). Voice search SEO is of course the process of optimizing digital content and webpages to assist in ranking, or being served, for voice search results.
Voice Search and Smart Speaker Adoption
Truly it has been the explosion of smart speakers, namely Amazon’s Echo and Google Home devices, that has brought voice search seo into the mainstream. The rapid expansion and diversification of affordable smart speaker options has led to increased adoption. ComScore reported smart speaker penetration among U.S. Wi-Fi households reached 11% in October 2017 and has likely hit 15% after the 2017 holiday season. 15% market penetration often represents a critical mass at which technology adoption can really accelerate. Smart speakers haven’t been around that long, but the forecasts say that there will be more than 100 million smart speakers installed in the United States by 2024, or roughly 30% of the population.
In addition to smart speakers and Siri, you can also use voice search on Google search (both desktop and mobile), the Google Assistant app, Cortana (both desktop and mobile) and on the Bing app (mobile only). With the roll out of Google voice search to 30 new languages around the world, it is now available to over one billion people worldwide.
Who is Using Voice Search?
All the way back in May 2016 it was reported that 20% of searches on Google’s mobile app in the U.S. are voice searches and that share is growing (this number is likely much higher since this data surfaced). At the same time Bing reported 25% of searches performed on the Windows 10 taskbar are voice searches.
We can also see from the chart above that U.S. smartphone users who use mobile assistants definitely trends to the younger demographic (as expected) but use among all demos is surprisingly high. We can take that usage and apply the percentage of voice searches above for some very strong numbers for total amount of voice searches.
Where and Why are People Using Voice Search?
Mary Meeker’s Internet Trends Report from 2016 reports the primary reasons why people use voice commands are: it is useful when hands are occupied, to receive faster results, and to avoid the difficulty of having to type. The report also shows the primary settings for where people use voice command are: at home (43%), in the car (36%), on the go (19%), and at work (3%).
There is a recent study by Stone Temple Consulting about the continuing growth of voice commands with smartphones that gives further insight into the situations in which people use voice commands and the environments people are using voice search in.
We can see the situations from the Stone Temple study definitely correlate to the primary reasons and setting for voice usage within the Meeker report.
Even more interesting is how the perceived social pressure seems to impact people’s willingness to use voice commands as highlighted from the Stone Temple study. You can see that the percentage of people willing to use voice search decreases as the environments move from more personal, secluded environments, to more public ones.
However it seems that is now already starting to change. Stone Temple has now completed this study two years in a row and compared the results. The 2018 results show significant increases in willingness to use voice commands in various public venues. In some cases, the willingness has more than doubled. It will be interesting to see how this willingness transforms as adoption continues to grow and as it becomes the norm with younger age groups growing up with the technology.
4 Steps to Win in Voice Search Optimization
These 4 key activities will set you up for voice search SEO success. Prior to optimizing for voice SEO, you first want to ensure you are optimized for Mobile SEO as voice search is an extension of mobile (and you need to ensure you’re winning at mobile anyway 😉
1. Use Voice Search + Explore How Your Target Audience Uses Voice Search
The first step in voice search optimization is pretty simple–read the official user guides on how to use voice search then start using it for your everyday searches. Apple has a guide on ways users can use Siri to guide them to personal information on their phone and information on the internet. Google and Cortana also provide similar guides.
In addition to familiarizing yourself with voice search and your query results, ask yourself “How are my customers and target audience using voice search?”
You can see below, again from the Stone Temple research, these different age segments are comfortable using voice search in different environments. Who is your target segment? How tech savvy are they? Review the environments they are using voice search in and explore what their voice searches may be within these environments and how those may relate to your business. What results are populating for those search queries right now? How can you compete and insert your content within those searches and conversations?
The ever-elusive and sought-after older Millennial segment (25-34 year old) is the leading segment within a majority of the environments below. This is a key market for many businesses. What would be some of their voice searches at home with friends or on public transportation? Would these searches be important to your product or services? Should they be?
2. Explore Long-Tail Keyword Opportunities for Voice SEO
A very exciting opportunity coming to light with the adoption of voice search is the expansion of new keywords and specifically longer-tail keywords. From the simple conversational nature of voice search the length of queries expands. This is a great opportunity within voice search seo as it opens a larger playing field for businesses to target topics and keywords and be more precise in targeting.
Look back at the different activities and locations your target audience is using voice search from step 1. What voice searches would they be performing at that time? How can that relate to your industry? Start complying ideas and hypothesis into what your consumers would be searching for during these times. Don’t put any guardrails on your research at this point, just start asking your own questions. Use a voice to text application on your computer or phone so you can document real conversational voice queries of your own. Or if you are using Google Assistant, you can view your history of searches and commands (and the results).
3. Answer Questions + Target Intent for Voice Search SEO
The true nature of any search, voice or text, is question based. The difference now with voice search is instead of typing a short 1-3 word query people are asking full questions. When voice search just started rolling a few years ago we were already seeing very large increases in question phrases within search, like “who,” “what,” “where,” “when” and “how.” While we (still) cannot equate these specific searches to “voice,” not surprisingly these increases were seen overwhelmingly on mobile and tablet where voice search was adopted more quickly. Such question based phrases have a higher probability to occur in natural speech, rather than in typing.
The important thing here is to structure your voice search optimized content to answer relevant search queries (questions) on your topic/subject.
But how do we get started? You already have a good base of voice queries relevant to your audience from step 1 and explored and documented some long-tail keywords in step 2. Now let’s put some search volume data behind that initial exploration.
There a a few great tools to help gather questions. We will explore two of my favorite.
Answer the Public is a great tool to visualize questions on your topic and drill down into the data. It’s best to start with a simple and broad single word. In this example I used “refrigerator.” Answer the Public will create two visualizations. One based on questions starting with: can, where, which, who, what, when, why, how, are, and will. I like to print this visualization out rotating around on my desk and highlighting interesting or important questions to my audience.
You can also select the “data” section to get a larger look at each of the questions under each section. Click the “Image” button within each section for another visual like below. These may help to save and print out as the initial visual above can be a bit compact. While you are going through this data keep in mind your research from steps 1 and 2. Do any of your initial queries (or similar) show up here here?
Of course you also want to download a CSV of the complete data (also including preposition based queries) for digital documentation and use plugin to other tools for actual search volume data.
Next let’s hop over to MOZ Keyword Explorer. Again here you want to possibly start with a simple broad query, though you can go a little more defined with this tool and start with longer queries. After you plug in your initial keyword, click on “Keyword Suggestions” then “Display keyword suggestions that” “are questions.”
Here you’ll see some similar results as Answer the Public but you’ll get relevancy to your initial keyword and of course monthly search volumes (within Google organic search results). You can sort by Relevancy or Monthly Volume and start checking off keywords important to your audience, business or campaign. Then add the keywords to a “Keyword List.”
Once in a Keyword List you can get a good idea on a number of factors including: monthly volume distribution, SERP features (pay attention to the Featured Snippets specifically), difficulty, organic CTR and priority across the set.
Here you can also “Add more keywords” from your previous research on Answer the Public and from steps 1 and 2. Now you’ll get to see what search volume those initial queries and questions have.
Now that you have some question based keyword research it time to synthesize that data down into the degree of a users intent and mapping to current or future voice search optimized content.
When a person simply searches “refrigerator,” you aren’t sure whether they are interested in purchasing one or just doing some research. But, when he or she puts forth a conversational question to an intelligent personal assistant, such as “what temperature should a refrigerator be?” or “how long does a refrigerator last?” or “where can i buy a refrigerator?” then you’ve got a much stronger understanding of the user’s intent.
Match up the questions from the research with the degree of intent, from simple interest to ready to act, and align the research within intent buckets to current content and/or formulate concepts and ideas for future content. You don’t want to push consumers into a ready to act conversion, you want to lead them down their natural path, ensuring you have the proper content to support their decision journey each step of the way.
4. Be the First Result: Rank Number Zero with Featured Snippets
Quick Answers, Featured Snippets, or (a form of Featured Snippets) called Answer Boxes allow you to rank above the #1 organic result and also provide answers for smart assistant/speaker voice search queries. A recent Backlincko study found 40.7% of all voice search answers came from a Featured Snippet.
How how do you go about gaining Featured Snippets for voice search seo? First, use your SEO platform to discover keywords which you, your competitors, or other within your industry are currently ranking for Featured Snippets.
BrightEdge’s DataCube is a great place to start by researching your domain, competitors, or other publishers in your industry. After entering a domain, click on the Content Strategies / Universal Results tab, then filter by Quick Answers (BrightEdge’s terminology for Featured Snippets). Be sure to actively track your Quick Answers rankings to monitor performance. Export your rankings as well as competitors / industry domains for opportunities / keywords to target.
In addition to current rankings, there’s likely a whole host of additional opportunities to gain Featured Snippet listings. The great part is, you’ve already done the research for these in the previous steps in this guide. Long-tail, question based queries with specific intent are great to target and optimize for Featured Snippets. Plug keywords from your research list into your favorite tool to track if these queries already have Feature Snippets. Combine this research with the featured snippet rankings you’ve already pulled and start finalizing your keyword list to target.
Before we get into further details on how to format and optimize your pages based on your new target keywords, you first need to ensure you can compete. According to research from Ahrefs, 99.58% of featured pages already rank in top 10 of Google. So be mindful of the domains ranking on page 1 (or already garnishing the featured snippet) for your target keywords and make sure your site can adequately rank on page 1. But don’t fret too much about the need to gain position one. 70% of snippets still come from sites outside of the first organic position. So while it’s required the page is ranked in the top 10, it’s not required to be #1 to be featured.
How to Optimize Pages for Featured Snippets
Create Long-Form Content with Conversational Language
Google voice results tend to be very concise. In fact, the Backlincko study found the average voice search result is only 29 words long. So to optimize for voice search SEO, aim to make your answer snippets approximately 29 words long. However, long-form content tends to rank best in Google Home searches. Backlincko also concluded the average word count of a Google voice search results page to be 2,312 words.
As you are optimizing your current content and building new content, ensure pages are created with natural language in mind. You should already be writing natural language copy for your content, but it is extra important to keep this step in mind for voice search optimization. You can see from the stat below that almost 70% of request to the Google Assistant are expressed in natural language (duh?), and these requests will not match up with the typical keywords people type in a web search.
You want to organize your target pages in an FAQ style, so that you can maximize the usage of question form keywords. Structure your content in the form of short paragraphs, each having a question subheading, with a long-tail keyword included. And remember, like any other great page or post, it is not about answering a single question, but being authoritative on broader topics. Test out answering many similar questions throughout your page.
Structure Your Voice SEO Content Accordingly
Google loves numbers, steps, and lists. Depending on the query/intent, featured snippets many times will list the actual ingredients for recipes, number of steps to complete a process, tables, images etc.
Based on the topic you’re covering, maximize the use of these content styles:
- Bullet points
- Numbered lists for steps (especially for “how to” queries)
It is very important that when utilizing tables, bullets and steps for your content your use HTML tables and HTML list tags. This gives Google the indication your content is actually a table or list and allows them to pull that data directly into the featured snippet.
An HTML table is defined with the
<table> tag. Each table row is defined with the
<tr> tag. A table header is defined with the
<th> tag. By default, table headings are bold and centered. A table data/cell is defined with the
An unordered list (bullets) starts with the
<ul> tag. Each list item starts with the
<li>tag. The list items will be marked with bullets (small black circles) by default.
An ordered list (numbers or letters) starts with the
<ol> tag. Each list item starts with the
<li> tag. The list items will be marked with numbers by default.
Featured Snippets Advancements
Featured snippets are continue to adapt to user search behavior. Two new additions to featured snippets have emerged recently. The first allows the users to interactively select/filter a featured snippet result based on your specific situation, such as selecting your cellular carrier on the query “how to setup call forwarding” below.
The second addition is multifaceted featured snippets, which show more than one featured snippet that’s related to what the user originally searched for or multiple interpretations of that query.
In the example below with the query “garden needs full sun?” Google is able to recognize that it could be interpreted as “what garden plants need full sun?” or “what counts as full sun?” and offer featured snippets for both.
I believe both of these new additions will open up further opportunities to rank for featured snippets and allow users better search results. One thing the multifaceted featured snippets to further introduce is pushing traditional search results further down the SERP, but that’s a discussion for another post.
How to Rank for Featured Snippets Recap
- Ask your target keyword/question in the form of a subheading on your page
- Immediately follow the question with a one-paragraph answer, preferably around 29 words and/or utilizing images, tables, steps/lists and charts
- Elaborate further on the question in the article
- Repeat this process answering several topically related questions, and similar questions with different user intent, throughout the page
Voice Search Optimization 2018 Steps Recap
Start using voice search and explore how your consumers uses voice search in their daily routine.
Research longer-tail keyword opportunities and optimize content with natural, conversational copy.
Answer specific questions target content based on the degree of intent, from simple interest to ready to act.
Rank for Featured Snippets