We hear about voice recognition almost every day. We already see how the technology changes our behavior and the way we shop. Amazon Echo, Google Home and other smart speakers already leads to a around $2B value of shopping done today in US and UK. In latest predictions about the voice shopping phenomenon, we already notice forecast for $40B in 2022 in US and UK, just 3 years from now. These are mind blowing numbers. To my opinion, these numbers are too optimistic, considering the current challenges that the voice shopping experience has. But from my experience, once the challenge is set, it is just a matter of time till some corporate, startup or service company makes a significant progress in order to solve it.
So let’s review the top 3 challenges for making the voice shopping experience as an experience that we would all use on our day to day shopping activities:
- Discovery — people don’t use voice speakers for product discovery (Like the way they do on mobile / desktop) — This is a critical challenge.
If e-commerce companies and in specific Amazon can’t improve their recommendation engine and enabling users to get inspired and browse for more shopping ideas, the revenues from voice shopping would be limited.
One optional way to solve it is by building a (super) smart AI engine with recommendation system that is connected with the information that we share from our age and family status, to our education, social networks, calendar of future events and even our physical shape. This can help the voice shopping systems act more like a voice recommendation system and recommend to us on different categories or selected items per category. For example, if I have a wedding that I am planning to attend 2 weeks from now, The voice system can recommend me to buy a new pair of pants that fits my size, send me some pictures to my mobile device, to my TV, or to any relevant device that I am using, and by saying “Yes” it will order the item for me. This is just one approach, in which we connecting the voice with the visual. I am sure there are already plenty of ideas and suggestions on how to make this as optimal experience and I don’t know what would be the ideal solution. But what I am sure about is that this issue has to be solved and once it is, it would be a game changer in terms of the experience.
2. Search– When reviewing latest research, it looks like about 50% of all searches will be voice searches by 2020, at least according to the different reports that are published. Consumers search differently when using voice search compared to regular search on mobile or web, and the main difference is that consumers tend to be very expressive in order to get the voice device to return with matching results. This means that we need to adjust our websites or mobile search applications to support it. In other words, AI is needed again — we need to make sure that our NLP (Natural Language Processing) algos would include natural language phrasing into textual content. This would help to deliver better search results. One example is how we enrich our different product pages with the relevant voice search terms for these products, so it will increase the chances to match between the search query to the searched item.
This content should be very focused on natural language, and how we really speak in the day to day. In many cases this is very different than the content presented on the shopping websites today.
3. Privacy — As Internet users, social networks, Facebook, Instagram and more, we know that we are giving up on some level of privacy in order to improve our own personal experience online. The same goes for online shopping.
However, having a voice device at our home or car, that in theory can actually record personal conversations, meetings and any kind of data that can be used by someone else is a not easy concept to live with.
If companies want to increase the usage of voice devices in general and voice shopping in specific, they need to come up with a smart solution that keeping only the relevant data for consumer behavior, or learn from the data without saving it on any other place. Yep, this challenge is hard, but as the other two I mentioned, I think is a clear must before the majority of the consumers would accept such a device in their home as part of their daily habits.
Voice shopping is exciting. It looks like it is going to change the future in a way we still can’t even imagine. but in order for it to make such a revolution, there would have to be a significant progress in the way it enables product discovery, search engine have to support it while enriching the product pages with regular language, and the issue of privacy have to be taken seriously.
Originally published on Medium