Over the course of 2017, there were three key trends that kept appearing again and again in the headlines -those of chatbots and voice search, Artificial Intelligence, and Augmented Reality. But are these trends really all they are cracked up to be or are they just a fad? Here is our run down of the top tech trends to watch in 2018 and if you should be doing anything about them.
All talk and no buttons of conversational UI
Conversational UI, or interfaces that mimic interactions with humans can be split out by either voice driven interfaces such as Siri and Amazon Alexa or messaging platform interfaces using chatbots like M in Facebook’s Messenger and Slack’s Slackbot.
Taking on chatbots first, brands are increasing implementing chatbots with lots of examples from fashion with M&S’s personal shopper service Try Tuesday, mortgage lending with Habito and even an NHS trial of chatbots to replace its 111 service in North London.
With so many examples of brands investing in chatbots, it is surprising to see a recent survey by Ubisend suggesting that only 57% of British consumers know what a chatbot is and 75% saying they have not spoken to a bot. This disconnect suggests that either customers do not know they are speaking to chatbots or they simply aren’t using the ones that are already in the market.
Interestingly the problem is the opposite for the other side of conversational UI in terms of voice interactions. According to Canvas8, up to 40% of UK households are expected to own an Amazon Echo device by 2018. And of UK smartphone users 37% of owners interviewed by Mindshare had used voice technology on their phones.
However, the technology and the number of brands using it has not quite matched customer expectations. We have seen ourselves, having recently completed an exploratory innovation project for a major broadcaster, that for some the technology is simply not there yet for a meaningful and worthwhile solution. And this is seen in the Canvas 8 research which found that 60% of users felt that if voice assistants could understand them properly and speak back to them as proficiently as a human interaction, they would use them more frequently.
Conversational UI is definitely an exciting space- and coming up with ideas on how to utilise the technology is easily done. The hard bit is finding out if consumers will actually use the solution and if the technology is mature enough to succeed. Our advice to brands in 2018 is to approach this work as an experimentation, learn fast and adapt to find the right solution for your customers, or your money will be wasted.
Artificial intelligence or artificial stupidity?
Artificial intelligence (AI) and machine learning have been making headlines in 2017, and this trend is likely to continue in 2018- but probably not in the way you would expect, with negative PR and poor executions plaguing this technology in the coming year.
AI is not a new thing, in fact, as pointed out in a recent talk by Pete Trainor the concept has been around since 1943. But it has recently gained pace, mainly because of the vast volumes of data we are currently able to collect and process – this is the “fourth industrial revolution”. And there are numerous examples of brands using AI in small ways, from Netflix’s personalised recommendations to Nest’s thermostat regulation. Philip Hammond has also recently announced a huge investment in driverless cards.
So what are we expecting in 2018? Firstly we expect to see a larger uptake in the use of machine learning to create personalised ad campaigns and personalised pricing. Currently only .5% of data is used to generate advertising or creative according to Forrester. This creates a huge opportunity for smart marketers to harness data-driven storytelling. But there is a catch – we are collecting more and more data, but our understanding and use of that data is mostly pretty poor. A machine can only be as good as the information we feed it – and poor data, or worse poorly understood data, can lead to poor creative execution.
2018 will likely see a rise in artificial stupidity and an increase in public scepticism – more customer service ‘humans’ cleaning up after chatbots and driverless cars making errors. But, it’s not all doom and gloom, we have an opportunity as an industry to start collecting data in a better way, supported by the upcoming GDPR deadline. And this technology will need errors and mistakes in order to learn from them and create a more robust framework for the next generation of machine learning technologies.