“Google, what kind of welder should I get?”
“Siri, where is the closest excavator dealer?”
Voice search isn’t traditional search executed on a mobile or voice-activated device. This increasingly popular search format impacts the way we structure queries and our search intent. For marketers, it impacts the content we create, the way we optimize it, and the appearance of search engine results pages (SERPs). Search engine optimizers (SEOs) are accustomed to an ever-changing algorithm and evolving search behaviors. Today those behaviors are changing in part because of a plethora of devices on which users are looking for information. When we think of voice search, we typically think of queries like, “Mexican restaurants near me” and “Alexa, order more toilet paper.” But today, voice search is leading to an increase in more complex and nuanced questions like, “Am I saving enough for retirement” and “Where should I live?”
Research indicates that not only is voice search on the rise, but personalized search is growing, too. We already know that consumers trust the advice and experiences of friends and family, and that reviews are a critical component of the purchase process (and your brand’s SEO strategy). But as devices that strive to meet our every need become more prolific – think Alexa, Siri, and Google Home – so too does the belief that all search results should be relevant and tailored to us.
According to Google, mobile searches with the phrase “__ for me” have increased 60 percent in the last two years, and mobile searches with the phrase “__ should I __” have increased 80 percent in the same time period. Questions like “How many credit cards should I have,” or “how much is my car worth” are becoming more and more common.
Is the intent of these questions different than, say, “car value calculator,” or “credit card tips”? Maybe. Maybe not. But the change represents a shift in how we approach search, and the content and context we expect to be delivered in a SERP. This has significant implications for marketers, especially at a time when more than half of professionals feel that they lack the technical know-how to harness big data. Users understand that their online habits are being monitored, and in exchange they expect to see that user data translated into relevant and frictionless digital experiences.
Voice search by the numbers
Sales of voice-activated search devices continue to grow. Alexa and Google Home dominate the market with 95 percent market share. As more users bring these devices into their homes, we’ll likely see other evolutions to search habits.
2. Google’s speech recognition and accuracy (English) is now 95 percent.
3. Search Engine Watch reports that Google voice search increased by a factor of 35 between 2008 and 2016.
4. According to a tweet from Purna Virji, more than 25 percent of Windows 10 desktop taskbar searches are voice search.
5. 43% of mobile voice searchers choose voice search because it’s “easier than going to a website or app.”
The last stat is particularly interesting. And while it might concern content creators and marketers alike – “But I NEED you to come to my website.” – it simply represents the need to shift the way you optimize content. Common sense tells us that searchers looking for detailed, in-depth content to help them make business decisions likely aren’t using Siri or Alexa to read them technical content. But like many other B2C trends of yesteryear, the habits we use in our personal life tend to bleed into the professional space, creating foundational shifts in the way we communicate, and as such, do business.
Tips for voice search
So how do we – as content creators, SEOs, and marketers – ensure our content is evolving as search trends evolve?
1. Look at your data. What does mobile traffic to your website look like – is it the predominant device, or is its share increasing over time? How do you rank in mobile versus desktop, and what do your click-through rates look like from each device? If you monitor your website’s rank for critical keywords (and of course you do), look to see if you’re dropping, maintaining, or increasing in rank. Use a tool like Moz or SEMrush to get a handle on the SERP features present in the results of your priority keywords. This will help you determine those keywords where you have the best opportunity to rank for position zero, or the featured snippet in a result. This is almost always the result that virtual assistants provide.
2. Look beyond Google. Yes, it’s the number one search engine on the planet. But keep in mind that many popular virtual assistants pull their results from other search engines. Siri provides Bing results, while Alexa uses a combination of Bing and your Amazon search and purchase history (remember that stat about Amazon and market share?). And of course, Google Home uses Google data.
3. Pull out your personas and hone in on the challenges your audience faces – could these be translated into personalized searches? Perhaps that’s not the case, but this trend in personalization reminds us that we need to understand the wants and needs of our audiences. What drives them to consider new products or technologies, or to invest in new equipment? Lowering costs or improving productivity is table stakes – we need to help our audience understand how our products and services make their lives easier.
4. Not working with defined personas? Think about the demographics of your audience, and the devices they may be using in their daily lives. For example, CMI reports that young people are generally better at voice search. Adults tend to use truncated queries – think “women’s running shoes” – and younger users lean on natural phrasing, like “what’s the best running shoe for marathon training.” Try to put yourself in your audiences’ shoes when optimizing content for search.
Ready to dive into the search traffic to your website? Here are a few terms to know as you start investigating and optimizing:
People also ask (PAA): In some SERPs, you’ll find a box labeled “People also ask.” These are other questions similar to your original query – they’re Google’s way of helping you refine your search further by including related questions that other Googlers have asked. It may also be a ploy to lengthen your search experience.
Featured snippet/position zero: Featured snippets appear as an expanded result in a box at the top of the SERP. They include the page’s title and URL, content pulled in from the page – generally longer than your typical description – and sometimes an image or graphic. The content is often in list or bullet form but can also be in paragraph form. It’s also referred to as “position zero” because it’s not technically the same format as a traditional result, and is often an expanded version of a result on page one.
Entity Authority: According to Mozzer Brittany Mueller, entity authority measures how reputable or credible your business is. Three primary factors are: how well-known your business is locally, whether the public has positive things to say about your business, and online engagement with your business. This is not to be confused with Domain Authority, which uses over 100 data points to assign a score, but focuses primarily on a site’s link profile, including number of linking root domains and the trustworthiness of those sites.
Machine Learning: As it relates to online search, machine learning is a step beyond artificial intelligence – a machine programmed to make decisions like a human begin to learn and evolve without additional programming. According to Albert Gouyet at Search Engine Watch, Google’s RankBrain is an example of a machine “that evaluates the intent and context of each search query, rather than just delivering results based on programmed rules about keyword matching and other factors.” In other words, the algorithm is trying to understand what the user is truly searching for, and often serving results that contain totally different keywords as the best possible answer.
We know that as voice search grows in popularity, it leads to an increase in natural language and long-tail search terms and changes the way we optimize content. But it also creates an expectation of relevancy that requires a deep understanding of your audience, their wants and needs, and their consumption habits. While mining your customer data for this information can be cumbersome, it will lead to quality content, better user experiences and ultimately, increased conversions.