Artificial Intelligence (AI) has arrived and many of us have 30 or 40 interactions with it a day, often without realising. Your Google search or Facebook feed; the ad that pops up on your computer that seems to know what you want; your Twitter trends; Siri and Google Home.
Gartner proclaimed AI a megatrend of 2017. McKinsey claimed that companies invested between $26 billion and $39 billion on AI in 2016. And Harvard Business Review found that 75 per cent of senior executives surveyed “believe that AI will substantially transform their companies within three years”.
For marketers, it’s a big – and complex – opportunity. AI’s ability to ingest customer data, find patterns within it and then use it to market to those customers – all without human intervention and in real time – was a pipe dream a decade ago. As was its capacity to determine the optimal combination of headline, image and body copy within a digital advertisement, and the right online media to place it in, to deliver the best results.
As digital marketing agency iPullRank said recently, “machines could potentially learn so quickly that you could remix, reuse and adapt content almost instantaneously, test the content you’ve created, and learn whether what you’ve created will be an improvement over your previous campaign or whether you need a different approach”.
All of this not only optimises a company’s marketing spend, it saves a huge amount of time, as marketing staff do not have to physically do the work. A Forrester study in mid-2017 showed that 80 per cent of retail marketers believed AI would revolutionise their role.
IBM believes in AI marketing so much it is using its own well-known AI machine, IBM Watson, to do its content marketing to highlight its role as a marketing technology leader.
US business Farmers Insurance Group is using AI to create streamlined claims processes. Rather than wait for an assessor to determine the damage and repair of a windshield, for example, the person claiming links up with the assessor via video call and shows images of the damage, which are then put through AI image-recognition software. The aim is to reduce claim times from days to hours, meaning happier customers and a strong point of difference for the brand.
Australian start-up Red Balloon has pivoted its business around AI marketing by not only implementing the AI software Albert but also buying the reseller rights to it here.
Albert claims to be the first full autonomous digital marketer, helping to move customers from one stage of their purchase process to the next based on its knowledge of their engagement with the brand. The software can identify relevant audiences and target them; test and optimise everything from keywords to creative combinations to budgets; plan, buy and restructure media plans 24/7; optimise the marketing mix; and uncover insights to act upon. In essence, it optimises and executes your marketing campaign.
Albert’s creators say its AI is best used in conjunction with human marketers. It sees Albert’s core skill as marketing optimisation and creativity as the domain of the marketing team. However there have been some terrific examples of AI’s involvement in creativity.
In 2016, a team from ING and Microsoft set out to use AI to create the next Rembrandt painting, 400 years after the artist’s death. They 3D-scanned 346 of his paintings and let AI’s deep learning capabilities create a new Rembrandt, from choosing the subject to creating the image and painting it.
Last year Hollywood used AI to create the trailer for sci-fi thriller Morgan. The studio ingested IBM Watson with 100 movie trailers for thriller horror films and let AI do the rest, picking the 10 scenes from the movie for the trailer.
Australia’s National Basketball League has followed other global sporting leagues in tapping into WCS Sports, an Israeli tech company that produces real-time automated sports highlights via AI. Great content such as slam dunks are posted on the NBL’s social media feed after being picked up in the live feed of the game.
In all these instances, however, AI needed to learn from existing material. It is not yet able to create something unique from scratch – a new sound, a new visual, a new concept. As in the Albert example, the combination of AI and human marketing is producing the best results.
AI’s marketing abilities will continue to evolve, particularly around voice conversations. Google recently unveiled an AI digital voice assistant able to book a haircut and a dinner reservation. For marketers, this raises the question of whether they’ll have to drive preference for their brand by marketing to the AI software – for example, someone asks the AI to order milk but the AI has to decide the brand.
Many companies have now made machine learning technology available to marketers, and the opportunities will only heighten as algorithms, data processing power and cloud storage improve.