Voice search in e-commerce: A view of the current situation and future of this new revolution (episode 1 of 3)

With the ubiquity of smartphones and smart speakers, voice is emerging as a powerful new way to interact with technology. This first episode of our series examines the growing impact of voice search on everyday habits and its potential for e-commerce. From simple voice commands to interactive conversations, voice search is redefining expectations for speed and seamlessness. But how is it being adopted in the online shopping journey? What challenges and opportunities does it present for retailers? Let’s dive into this fascinating world where innovation and customer experience converge.

What is voice search ?

The arrival of voice in search

With the widespread use of smartphones, whose keyboards don’t always make typing easy and whose screens are smaller than those of tablets or computers and display less information, new modes of interaction have emerged. In simple terms, smartphones began to talk to us when they first transformed themselves into GPS devices. They also became task and agenda managers, and finally developed into assistants to make our lives easier (making calls, reading/sending SMS messages, answering questions, etc). It is now possible to interact with your phone without using your hands at all, particularly during tricky moments such as driving a car.

Intelligent personal assistants and, increasingly, automation solutions in the connected home, further enhance the appeal of voice for the management of our daily tasks.

Since then, it has become natural to switch from using a simple voice command (to perform tasks and actions) to searching for information directly on a smartphone without having to touch the screen or type words. Nowadays, we can use voice - the oldest and most natural method of expression - to receive answers to our questions increasingly quickly. We have progressed from «simple» voice commands to a mode of conversation with the connected objects around us that is interactive and intelligent.

Until now, user searches have always consisted of a series of keywords, often out of context, typed into Google’s search bar. However, the arrival of voice is disrupting habits: instead of a few key words being entered, entire sentences are being spoken aloud. When people speak, their queries are naturally longer and more contextualised; similar to the sentences used in the natural everyday language of human dialogue.

Google’s algorithm reinvents itself every day and offers more and more artificial intelligence to support the user experience. From finding the right route or the nearest shop through to booking a train ticket, conversational search will change the way we interact with technology. Today it is no longer a case of people learning the language of computers; computers are learning the language of humans.

For e-commerce businesses, this means coordinating all the internal search tools that allow customers to access information on their sales sites, as well as all the devices that are now available to them, including those that use voice. As voice gains recognition for being one of the essential new tools for finding products, we are reaching a turning point.

How is voice search used currently?

Of course, the advantage of voice search is to simplify our lives by saving us time (dictating is on average 3 times faster than typing) and to avoid restricting us in situations that require mobility, such as driving or cooking.

More and more connected objects have no keyboard and are therefore ideal candidates for voice interactions. This is particularly true of connected TVs in the US, the usage rate of these is growing faster than any other voice search devices.

There are of course uses that are more common than others, such as searching for places or addresses when travelling by car, for which voice search was originally developed. Then there are searches on news, weather, and also for recreational content such as music or videos. It should be noted that there is a very low rate of external voice searches for products. In other words, voice assistants are still currently struggling to prove their usefulness during a shopping trip.

Voice is first and foremost software

Connected speakers are widely available and contribute significantly to the widespread use of voice. Last year, the US market grew by nearly 40%. In May 2018, there were 55 million users. By January 2019, there were already more than 66 million. According to UK analysts at Juniper Research, 8 billion voice assistants were used worldwide by 2023, with a rate of growth of 25.4% between 2019 and 2023 !

A connected speaker is also a tool shared by several people in a family, which further multiplies the number of potential users. These connected speakers incorporate a smart voice assistant, which is essentially a piece of software that can be installed and used on different types of device that possess a microphone or a loudspeaker (such as smartphones, speakers and other connected devices).

It’s important to make a clear distinction between the physical interfaces (i.e. the connected speakers which we speak to physically, such as the Amazon Echo, Xiaomi speaker or Google Home and Mini) and the software which renders them intelligent via the algorithms that interpret our language – such as Amazon’s Alexa, Google’s “OK Google” and Apple’s Siri.

The battle for voice intelligence is on

At the beginning of 2011, Apple made voice available on the iPhone by creating the “Siri” algorithm. It was joined one year later by Google, with the “OK Google” assistant and then in 2015 by the world leading retailer Amazon, with its Alexa software.

Although Siri had the historic advantage, Google and Amazon quickly caught up, with investments and deployments across the world. For example, Google’s voice assistant is already integrated on one billion devices *1, 10 times more than the 100 million devices connected to Amazon Alexa, which is still heavily reliant on Amazon Echo speakers.

There is currently no direct competitor to the offering of Amazon’s voice assistant, however. Alexa possesses “skills”, or features that can be activated according to what you want it to do. In January 2019, Alexa had over 60,000 features with the potential to be activated, against only 4,253 from its direct rival Google Home (which calls them “actions”).

This is why, in January 2018, Amazon had 72% of the global market share for connected speaker sales worldwide. The more intelligence there is in a speaker, the more interest there is in using it. Amazon is unquestionably ahead in this, thanks to the quality of its APIs, which allow third party businesses to develop new skills easily for its range of connected speakers.

Amazon “skills” – or Google “actions” – are in fact downloadable apps, which allow brands to offer a service to their customers, using the connected speaker as a point of entry. The aim is to develop new applications using frameworks and by exploiting the technological and functional capacity of the voice assistant - and of the operating system specific to each brand (a very similar model to the development of mobile apps).

In France, Google dominates usage at almost 32%, just in front of Apple’s Siri (31%) and Microsoft’s Cortana (21%). Amazon’s Alexa (6%) is lagging behind, almost certainly because of its late arrival into the French market.