By AI Trends Staff
Corporate marketers are using AI to more deeply analyze the customer experience, and to augment analytics with new approaches and new tools. Here is a review of recent trends in the use of AI by corporate marketers:
Corporate marketers surveyed in August 2019 indicated high interest in rolling out more AI capability, according to the CMO Survey as recently reported in Forbes. The corporate marketers surveyed had increased their use of AI and machine learning in marketing toolkits by 27% over the previous six months. The surveyed marketers projected a 57% increase in use of the AI tools in the coming three years.
Companies with $1 billion or more in revenue and high rates of their sales via the internet were projected to spend more on AI, and they are able to hire needed data scientists to help engage customers. Adoption rates of AI by marketers varied by industry, with the highest projections in transportation, technology and education; the lowest in manufacturing, mining and energy.
The top applications of AI by marketers focus on deriving more value from customers, namely: AI for content personalization (57%); use of predictive analytics (57%) and targeting customer decision-making (50%). Other applications were to optimize advertising and media buying, fine-tuning marketing content and timing, and implementing conversational AI in customer service.
The founder and director of The CMO Survey is Christine Moorman, a Professor of Business Administration at the Fuqua School of Business, Duke University. Her advice to marketers for what to consider when pursuing AI projects included points on using analytics and finding talent.
Christine Moorman, founder and director of The CMO Survey, and a Professor of Business Administration at the Fuqua School of Business, Duke University
“AI is only valuable if marketers use resulting marketing analytics,” she stated. The CMOs reported using marketing analytics to help make decisions 39% of the time, meaning they were mostly unused. It was an improvement, with 29% reporting it was more than they did in 2013. Moorman challenged them to do better.
“Companies will have to work harder to build AI-driven data and decisions into their standard operating procedures so that they systematically produce a real payoff,” she stated.
Finding the qualified people to make the AI team is a challenge for marketers as it is for managers throughout business. “A gap in talent makes the use of AI challenging,” Moorman stated. Only 2% of respondents said their companies have the right talent to leverage the marketing analytics needed for AI applications. The options are to get lucky in hiring, buy services from a large technology firm, or adjust the AI plan to fit the available talent, she suggested.
Customer Experience Can be Enhanced with Data Unification Tools
Customer experience (CX) can help a company grow when successful, and be a high source of risk when it does not work well. Insights from data are a primary way CX can be improved; however, customer behavior can be chaotic, so interpreting the data is a challenge. “The rules are undefined and the success criteria are ambiguous. CX is the nightmare dataset for an AI developer,” stated Will Thiel, co-founder and principal product architect behind Pointillist, in a recent piece posted on the Pointillist blog.
Will Thiel, Co-founder and Principal Product Architect, Pointillist
A successful application of AI in customer experience relies on three building blocks: data unification, real-time delivery of insights, and business context, advised Thiel, whose company provides solutions for data unification, more effective customer segmentation and more personalized engagement using AI.
A new generation of data unification tools makes the task of data unification reasonably priced, fast and fairly pain-free. “The tedium of pulling together dozens of data sources is now just background noise,” he suggested.
To deliver insights from every touch point of a customer, customer journey analytics platforms are offering many API options and development kits to help create touchpoint integration.
The way customers interact with a site is distinct for each company. “Customer journeys are as unique to individual businesses as fingerprints,” Thiel stated. For AI to have value, it must know the significance of each customer behavior event in shaping customer behavior. It must know which key performance indicator is affected by customer behavior, whether related to revenue, profitability, customer lifetime value, customer satisfaction or other factors.
“With proper business context, an AI can find touchpoints and tactics which actually shape the customer behaviors behind the business’s primary measures of performance,” Thiel suggests.
A company making good use of AI in analytics is Sephora with its Visual Artist product, he suggests. Visitors can try on cosmetic products such as lipsticks, eyeshadows and highlighters to match their skin tones. ‘Using AI, the tool can map and identify facial features and apply the product to the user’s face,” Thiel stated. “Sephora has thoughtfully considered the entire customer journey. Sephora has surged ahead in AI usage.” Naturally Visual Artist ties into the company’s inventory of products seamlessly, able to make personalized recommendations and offers in real time.
Augmented analytics is helping to advance the field further. Gartner rates it in the top 10 of technology trends in data and analytics. Augmented analytics is defined as the use of enabling technologies such as machine learning and AI to assist with data preparation, insight generation and insight explanation, suggested a recent account in RT Insights. Related tools can assist expert and “citizen data scientists” by automating many aspects of AI model development, management and deployment.
Augmented analytics can be used to help create algorithms to help the users do something they could not do without the tools. In one example, a bank had been targeting older customers for wealth management services. Using augmented analytics, the bank found that clients aged 20 to 35 were likely to transition into wealth management, so also made good targets.