With a recent forecast from IDC estimating that worldwide revenues for artificial intelligence (AI) will total more than $156 billion in 2020, the writing is truly on the wall. With a global pandemic disrupting customer service at every turn, it’s time to pull AI from the pilot projects and use it to optimize the consumer experience.
“AI-powered tools and capabilities allow us to understand and communicate with customers like never before,” said Hayley Sutherland, Senior Research Analyst, Artificial Intelligence Software Platforms with IDC’s software market research and advisory practice.
A researcher who focuses on image and video analytics, voice and speech analytics, biometric analysis, and affective computing, Sutherland joined ITWC President Fawn Annan in November 2020 for an installment of CMO Talks, a podcast series presented by ITWC and IDC to address pressing marketing challenges. Their discussion centered on how organizations can leverage AI tools to enhance brand perception and increase Net Promoter Scores (NPS).
Natural Language Processing
Sutherland was particularly excited about advances in natural language processing (NLP), specifically the use of AI to understand natural human communications. “The really cool thing about some of these big leaps in natural language processing is that we’re better able to understand what people are looking for,” she said. “Not only can we start to understand what our customers said and what they searched for, but we can also understand why they communicated with our brand and what sentiment was behind that decision.”
Sentiment Analysis
Sutherland described text-based sentiment analysis (also referred to as emotion AI and opinion mining) as one of the most interesting applications for CMOs. “That’s using natural language processing to understand how someone feels, which has many implications for how they feel about our products and brand,” she explained. “By using AI to understand what that looks like across the customer base as a whole, as well as for an individual, we can understand how to better connect with customers and to get them to better connect with our brands.”
In answer to a question from Fawn Annan about how AI-powered search increases marketers’ conversion rates, Sutherland related an anecdote about typing the word ‘bat’ into a product website’s search bar. Based on current trends, the time of year, and her personal context, including what she had bought in the past, the system should be able to figure out whether she is looking for a baseball bat or a Halloween decoration. “So now I have found what I need,” she said, “and I didn’t abandon my search, or go to a competitor because I instantly got what I was looking for instead of having to sort through a mishmash of Halloween decorations and sporting codes.”
Voice-based Search
According to Sutherland, voice-based search capabilities are becoming more accurate and consistent with the rise of home assistants such as Siri, Alexa, and Cortana. “People are becoming more comfortable with using voice-based search and voice-based commands,” she said. “It’s another mode to make it easier for people to find what they want, and with COVID-19, there’s the added benefit of having a touchless experience.
Visual Search
From there, Sutherland moved on to consider AI-powered visual search, or the ability to use AI to find either specific images or videos. Her vote for the most interesting application for the CMO audience is the use of AI to find similar images, exemplified in Amazon’s StyleSnap, an AI-powered feature that lets users snap a picture of a fashion look they like, upload the image, and receive recommendations for similar items on Amazon.
Without getting too deep into what she described as the “technical weeds”, Sutherland summarized how metadata improvements have improved AI powered visual search. “In the past, the only information a computer had about an image was text-based,” she explained, “but in recent years, AI has begun to analyze and understand the image content. Because of that, we can start to embed related concepts, related terms, and even webpages within an image file.”
The Benefits of AI
Sutherland credits advanced AI technologies with helping consumers find products more quickly. “Being able to offer voice-based services is something that customers have come to expect,” she says. “It elevates you as a brand, and also has a major impact on operations for customer support – especially in these times.”
After providing a summary of the key players in speech to text software, she referenced a recent IDC survey showing that voice-based interfaces were at the top of the investment list for technologies that support contactless experiences. Her advice for marketers when considering AI technologies is to know how they plan to use them. “It might seem a little obvious,” she said, “but it’s so critical to understand how the use case can be tied back to metrics that matter to the business.”