print logo

AI is there for the customer

Thanks to AI, productivity will increase even more dramatically because it can process a large number of customer requests faster than humans.
Tim Cole | 24.02.2021
AI is there for the customer © Pixabay
 

"Your call is important to us," a taped voice tells resigned customers who are endlessly on hold, waiting to speak to a human agent. This is where AI can help companies improve the quality and consistency of their service and convince customers that you actually care about their concerns.

AI could change customer service as much as the telephone once did. Before that, after all, companies handled customer queries by mail or face-to-face visits. The phone helped agents become more productive.

Thanks to AI, productivity will increase even more dramatically because it can handle a large number of customer queries faster than humans. This has become even more important as communication channels have multiplied: There are more options available to the customer than ever before - email, mobile messaging apps and social media - and it is always the customer who decides which channel to use to contact the company.

Consumers have become accustomed to dealing with automated services. Surveys show that about 40 per cent of American internet users prefer to use digital customer services rather than talk to someone on the phone: self-service, it seems, is perceived as the best form of service.

Ocado, a British online grocer, receives about 10,000 emails a day from customers and uses AI to detect their mood. This means the most urgent complaints are prioritised first and routed to agents with the right expertise in the relevant area.

"As with other applications of AI, it's about making people more efficient, not taking them out of the process altogether," says Paul Clarke, Ocado's chief technology officer. By 2021, the proportion of customer service interactions handled entirely by AI globally is expected to quintuple to 15 per cent, and by 2019 at least 40 per cent of those interactions will involve some element of AI, according to market researchers Gartner.

Virtual agents are on the rise. About 30 per cent of companies now offer their own "bots" that can answer questions and solve problems, although their reach is still less than that of a human. Many of them use AI; they are trained on logs and transcripts of previous customer interactions, and the more data they get, the better they can solve more complex queries. Such bots allow companies to handle many more enquiries without having to hire additional staff.

 

By their voices you shall know them

AI will also improve the knowledge, performance and speed of customer service agents. Some companies are experimenting with voice printing - technology that recognises customers' voices and alerts agents when a caller tries to impersonate someone else.

This will be particularly helpful in financial services. An Australian bank is experimenting with a standalone smart voice-controlled speaker to listen in on its agents' loan conversations. If the agent forgets something or makes a mistake, colleague Bot steps in. Other companies are using AI to suggest answers to customer queries that a human agent can approve or adjust before sending. Last year, KLM, the Dutch-flagged airline, doubled the number of text-based customer queries it handled to 120,000 a week as a result, and only had to increase the number of agents by six per cent to do so, says Dmitry Aksenov of Digital Genius, a company that helps automate customer support.

A few companies have already started offering AI-enabled services that listen to calls to assess agents' performance and send them real-time suggestions for improvement.

AI with emotion

Cogito, a Boston-based start-up founded by an MIT alumnus, has developed software that performs emotion and mood analysis on call centre conversations using natural language processing. It measures energy levels, pace, manner of speaking and other factors in real time to detect and interpret speakers' intentions so they can spot mistakes and make spontaneous corrections. For example, if an agent is speaking too fast, Cogito's software might suggest that they slow down or address the customer with a question. Using AI, Cogito determines an empathy score for each call centre employee, depending on how good the agent is at conveying empathy to the customer and successfully resolving complaint cases.

Cogito uses Artificial Intelligence to augment human intelligence, particularly emotional intelligence. "We use technology to help people be more human. Almost ironic, isn't it?" says Cogito CEO Joshua Feast.

Cogito grabs channel-separated audio from the phone and is therefore able to isolate both customer and agent speech patterns and conversational exchanges. During a call, the technology breaks down the conversation into millisecond increments and analyses over 200 different vocal non-verbal signals such as porosity, tone, tension, speech rate, vocal effort, turn and mimicry. These signals are then analysed and correlated with insights from millions of other conversations and with the help of artificial intelligence.

During a live call, Cogito's low-latency compute engine analyses hundreds of unique behavioural signals to deliver objective behavioural instructions in just milliseconds.

Cogito's customers include insurance companies such as Humana and MetLife. There, Cogito focuses on detecting "compassion fatigue". It looks at details like the agents' rate of speech and the words callers use. The AI can use this to detect emotions and assess whether the interaction is satisfactory. If there is a problem, the agents are encouraged to act more empathetically.

Even though agents know they are being monitored, most welcome such measures, according to Cogito, because they give them valuable feedback. It pays off for the company: a call centre's turnover can increase by between 30 and 40 per cent per year.

Marty Lippert, MetLife's chief technology officer, expects AI to bring a return on investment (ROI) of around 20 per cent in areas such as customer service and human resources. Most companies buy AI services from external providers, but firms with in-house technical expertise often prefer to create their own AI. For example, a team at Uber has built a system for handling AI queries via email, which is offered alongside the traditional phone number and sends the agent a ranked list of options for how to proceed, reducing the time it takes to handle a complaint by around ten per cent.

Please don't call us!

Gartner believes that the number of telephone-based customer service staff worldwide will fall by ten percent by the end of 2020. This would increase the workload of those who remain. But companies must be careful not to dilute their interactions with customers too much. The rise of virtual communication means they have fewer opportunities to build deep customer relationships, making customer service increasingly important.

Services that make customers' lives easier will generate more customers providing more training data to make AI systems smarter.

Robotic process automation (RPA) as a gateway technology for intelligent process automation will also play a huge role in marketing and sales in the future, such as performing credit checks, updating customer data, assigning clerks and personalising content. Every brand interaction a consumer has, be it via an e-commerce website, social media or in-store, will be tracked and stored by a marketing platform in the future. So-called "predictive" solutions then dig deep into these insights to determine the logical next steps and predict actions that customers with certain consumer profiles will take. The result of these analyses can automatically trigger actions by the company through various channels, such as email, mobile devices and the web.

There are now powerful predictive tools such as Salesforce.com's Einstein or Atomic Reach that can help understand what content will resonate well with a target audience. Concured, which sees itself as a "listening tool", uses artificial intelligence to analyse consumer behaviour in relation to content and shape future content marketing. Such tools are critical in creating and executing personalised marketing strategies. To create truly personalised buying experiences for consumers, marketers need up-to-date insights into individual audience members and then be able to generate and deliver the unique content accordingly. When consumers receive this tailored content in near real-time, it greatly increases the efficiency of marketing teams' work.

In marketing, different types of predictive solutions present themselves. Predictive mail solutions can deliver targeted recommendations to consumers via mail by analysing their past purchase history, email engagement rates and the historical affinities of similar consumers. Other types of referral emails include abandoned cart campaigns, post-purchase campaigns and abandoned browse campaigns. When a consumer visits an online website, predictive analytics views can be used to create a personalised shopping or browsing experience by placing certain recommended products above the fold or generating a sidebar of "frequently purchased together" items.

Knowing what the customer will want

In the meantime, the buzzword "predictive marketing" is beginning to catch on in professional circles. Those who already know tomorrow's customers today have the chance to use their advertising budgets more efficiently. Knowing when a customer is ready to buy will be decisive for marketing success in the future. However, this requires a sufficiently large database of customer and sales data. The decisive factor here is the intelligent merging of data from different data sources in order to identify sales opportunities and to use the insights from data analysis to identify target groups that are ready to close a deal.

Predictive marketing is primarily about creating so-called personas, which statistically possess characteristics that are relevant for closing. Through the data-driven identification of closing potentials based on historical data, advertisers achieve a much more precise target group description and identification. Data must be collected and bundled across different channels, taking data protection into account. This makes it possible to draw conclusions about potential target groups in the individual channels.

A prerequisite for successful predictive marketing is a well thought-out strategy. This must not only include the usual demographic characteristics, but also things like times of day and user behaviour. This is the only way to create an overall picture of the customer and obtain answers to key questions such as through which channels the statistical personas should be addressed, how marketing processes can be adapted to personas and which algorithms are necessary for the prediction.

As always in AI, the be-all and end-all is a sufficiently large and well-maintained database. But where to get it from if you have not systematically collected customer data in the past and stored it in a retrievable form? In such a case, AI-based personalisation platforms enable simple and cost-effective, individualised customer experiences across all contact points such as website, app, email, POS system, IoT device and call centre. Providers such as Nosto or Dynamic Yield use machine learning and AI to personalise and optimise the so-called "customer journey", i.e. the journey of each individual customer to the product in real time. Based on self-learning algorithms, brands can directly test what works well and what does not. Customers then receive personalised offers in real time or purchase recommendations by email. "Personalisation will soon be omnipresent," says Liad Agmon, founder and CEO of Dynamic Yield, "because it is an absolute must in marketing!"