See Algolia in action
Tailoring search to maximize click-through and conversion rates
Sorry, there is no results for this query
For years, people have made sweeping predictions about how voice search would soon outpace all other searches. But, as the dust has settled, we’ve seen much more modest growth than expected. Still many users are gravitating toward voice search, with voice-first devices and virtual assistants like Google Home, Alexa, and Siri becoming staples in many homes. It’s clear voice search plays an important role in the evolution of search, and it is poised to grow as more consumers get comfortable with voice search.
But this changing landscape has many sites and businesses wondering how voice search might affect them. Which voice search trends are here to stay? What is the reality of voice search today? How might voice search look like in the future?
The voice search of today can be traced back to some early predecessors.
Voice recognition emerged in the 1950s with software by Bell Laboratories that could recognize basic digits. From there, innovation and refinements came along, including the U.S Department of Defense’s DARPA SUR project in the 70s, the development of the Markov model in the 80s, and the advent of faster microprocessors in the 90s.
By the early aughts, Google created “Google Voice Search,” an application for iPhones that leveraged cloud data centers for processing power. Over the next several years, Google continued to innovate, fine-tuning their software with a wide range of examples of human speech and training data.
Voice-first technologies allow humans to interact with computer systems through voice, but early interactive voice response (IVR) devices did this first. Today, most modern users interact with IVR through call centers, where an automated system allows the user to make choices through voice or keypad. This technology first emerged in the 60s, as improvements to synthesize speech were sought to drive efficiency in call centers. More advanced IVR was particularly notable because it presented the first time voice and web had really been connected.
These technologies laid the foundation for the voice recognition and voice search technology we see today.
In a 2014 interview, Andrew Ng famously predicted, “In five years time at least 50% of all searches are going to be either through images or speech.”
Well, we’ve just passed the 5 year mark. So how does voice search stack up to the prediction?
Evidence shows voice and image search have not yet reached 50% of all searches. However, there is a lot of exciting research that shows how powerful voice search is and how users are integrating it into their lives.
What do all of these latest stats mean in practice, though? What are the trends that businesses and users should take note of? Well, for starters:
Although the growth of voice search has been more steady, measured, and modest, users are definitely integrating voice-first devices and capabilities into their lives. And, this comfort with voice search means a change in search behavior. In fact, 71% of users would prefer to search through voice than use a keyboard. Other research shows 40% of millennials research products through voice before purchasing. These statistics tell a compelling story: people are ready for voice search.
For years, searching on websites and search sites like Google meant putting a few keywords into a search bar. These days, modern internet users are moving away from the traditional method and moving towards more conversational queries.
As users speak to voice-enabled devices and assistants, they tend to use natural speaking patterns, rather than stilted keyword constructions. In fact, 70% of queries to Google Assistant are in natural language. This means queries include more phrases, complete questions, and common strings of natural language.
The speed and relevance of major players like Google, YouTube, and other prominent websites have upped the ante for sites across the web. Users now expect fast, relevant, and personalized results for every query. Therefore, when users utilize voice search, they expect to receive results flawlessly.
Unfortunately, despite high user expectations, the number one pain point of voice search is not getting the right answer to their query. User expectations won’t decrease anytime soon. Thus, it’s paramount for voice search to quickly reach the same levels of speed and relevance as more traditional, text-based search methods.
Many companies are already investing in optimizing their sites for voice search. Unfortunately, some overemphasize either voice-related SEO improvements or the programming of specific voice search skills in their strategy.
In reality, preparing the site for voice search is so much more. It requires understanding and prioritizing customer needs first so that the site search and content is tailored to natural language requests. Companies need to prioritize delivering a great search experience on their own websites and apps.
Voice search is not just confined to personal devices; many physical stores are deploying it to their advantage as well. Using voice-first devices, such as a voice-enabled kiosk in a brick and mortar location to help customers locate items, can ensure there’s always someone there to improve the customer’s real-life experience.
As the capabilities of voice search continue to expand, the way that customers express needs and the way companies meet those needs will change. Some ways voice search might develop in the future include:
Over time, voice search devices will begin to understand what customers might be looking for even before they search. With AI capabilities, the search systems will be able to learn customer preferences and tie in relevant context such as location, weather, etc. to make the most valuable product and content suggestions to users.
Ultimately, voice search will continue to be driven by how users choose to integrate it into their lives. Based on the strong adoption so far, it’s clear users see the value in a strong automatic speech recognition (ASR) and natural language processing (NLP). As this buy-in continues, companies will be incentivized to keep investing, improving, and offering the technology.
Similarly, voice search’s future depends on how thoroughly companies adopt it. They can choose to devote adequate resources to integrating voice for virtual assistants into their sites and apps, ensure the assistant ecosystem is interoperable, and build their own assistants. Or, they can take the easy road and just put a mic on their mobile apps or just plug into existing assistants.
As people become more comfortable talking to Alexa, TVs, our mobile phones, and just generally speaking to computers in front of others in public, more mundane “human” or repetitive tasks will be handed over to computers. This will happen in a way similar to how mobile banking has replaced physically going to the bank to make a deposit. Home devices can already perform tasks such as lowering or raising the heat or lights and turning on kitchen appliances. This functionality will continue to expand.
Of course, there are functions that voice search may serve that haven’t even been imagined yet. Imagine going to a home improvement store and getting recommendations for products and finding those products all through voice search. Similarly, purchasing a ticket at any event or attraction could easily be done with voice applications. Once the technology catches up with user expectations, the sky’s the limit for voice search.
Voice search is definitely here to stay. Make sure voice search is part of your greater strategy for well-optimized, intuitive site search.
Read more about preparing for the future of voice search in our ebook “The Next Tech Revolution Will Be Spoken.”