Artificial Intelligence (AI) has gone mainstream when it comes to customer interactions, according to a new report from the Capgemini Research Institute. More than half of customers (54%) have daily AI-enabled interactions with organizations – a significant increase from the 21% reported in Capgemini’s 2018 research on the subject.
The report, ‘The Art of Customer-Centric Artificial Intelligence: How organizations can unleash the full potential of AI in the customer experience’, reveals the factors that have significantly contributed to AI adoption among customers, including increasing customer trust in AI; an increase in human-like AI interactions; increasing customer concerns arising from COVID-19; and organizations stepping up their AI deployments.
COVID-19 has accelerated customer adoption of non-touch AI-based systems, such as voice assistants and facial recognition – a shot in the arm for AI adoption. Over three-quarters of customers (77%) expect to increase the use of touchless interfaces to avoid direct interactions with humans or touchscreens during COVID-19, and 62% will continue to do so post-COVID. This figure is even higher in countries such as Germany (73%) and Brazil (71%). The fact that touchless interfaces are becoming integral to the customer experience in a health and safety-conscious world is also recognized by organizations: 75% believe that increasing customer appetite for non-touch practices will persist even in the post-pandemic world.
Customers have significantly increased their preference for AI-only interactions; Kelly Anderson, Director, Data Science and Artificial Intelligence at Procter & Gamble confirms this saying, “I believe that customer expectations have evolved to the point where they almost expect for interactions to be AI. So, when you actually put a human in the loop, they are very pleasantly surprised and sometimes shocked. This clearly shows that chatbots / Natural Language Processing /AI are making progress and have evolved.”
From a sector perspective, Automotive (64%) and Public Sector (62%) stand out as the strong performers. The widespread usage of in-car voice interfaces explains the dominant position of automotive, in part. For instance, BMW, which has been deploying its own in-car AI based voice assistants for many years, plans to make them more natural, with gesture recognition or gaze recognition capabilities for its 2021 series[1].
Trust and human-like interactions have improved
Trust was identified as an area for improvement in 2018 and the latest research reveals that organizations have made great strides in this area. Over two-thirds (67%) of customers trust the personalized recommendations and suggestions provided by AI-enabled interactions. Moreover, close to half of customers (46%) find AI-enabled interactions to be trustworthy – compared to 30% in 2018, while the share of customers who say that they do not trust machines with the security and privacy of their personal data has dropped to 36%, down from 49% in 2018.
Customers also wanted more human-like AI interactions, and organizations have progressed here too. Overall, 64% of customers believe that their AI interactions are more human-like (compared to 48% in 2018). China (74%), Australia (72%) and the US (70%) lead in the percentage of them who believe that their AI interactions are more human-like. Organizations have been consciously trying to build human-like features into AI applications: 72% of organizations agreed that they are actively trying to make their AI interactions more human-like.
“Context-aware” AI use cases hold the key to dropping customer satisfaction levels
While customers have increased their AI interactions since 2018, their level of satisfaction has dropped. Overall, 57% of customers are satisfied with AI interactions, compared to the more than two-thirds (69%) who were satisfied in 2018. Additionally, 51% of customers say they will consider an AI experience to be positive if it provides a unique experience beyond their expectations.
The research found that more customers are satisfied with “context-aware[2]” use cases and receive greater benefits from them than those with the rest of the use cases. Examples include autonomous parking of cars, detecting fraudulent banking transactions, and making payments authenticated through biometric scanners.
Most organizations measure AI performance in customer experience with basic KPIs
Few customers have experienced AI in a way that far exceeded their expectations. This can be linked back to the fact that a majority of organizations (73%) only follow a basic KPI (key performance indicator) for measuring customer experience, which only looks at the number of customers served by AI interactions. Organizations must add measurement and feedback management into AI design and development, cites the report, so that AI systems can deliver their true potential of learning and improving over time.
The future of customer experience
Capgemini’s study from 2018 found that most organizations (93%) had less than 30% of customer interactions enabled by AI. Today, only 10% of organizations are at that low bar, with 80% saying that 30% to 50% of customer engagements are AI-enabled. According to the report, in two to three years’ time, the vast majority (80%) will have more than half of their interactions enabled by AI.
“Usage of AI for customer experience is clearly here to stay. COVID-19 has been a catalyst in moving organizations towards AI implementation, and changes in customer behavior mandated by the crisis have created a clear opportunity to scale AI implementations,” says Darshan Shankavaram, Head of the Global Digital Customer Experience Practice at Capgemini. “However, it is integral that businesses focus on using AI to delight their customers and create better interactions and experiences, rather than simply using it to address volume or as a technology innovation. Going forward, we expect to see customer satisfaction improve and their openness to using AI further along the customer journey increase.”
To read a full copy of the report and its recommendations, click here.