There is no hotter topic with CIOs than digital transformation. Going digital has become the primary mandate for IT leaders as companies look to leapfrog one another. When it comes to digital transformation, there is no single project that can take a company from where they are today to being fully digitized. Rather, digital transformation is a set of smaller projects that will eventually lead to the evolution of the company.\nAlthough projects can vary widely by industry vertical, many of them have one point of commonality: They are focused on improving customer experience as that becomes one of the key differentiators for businesses moving forward. What\u2019s critical for CIOs to understand, though, is that the focus shouldn\u2019t be on slightly improving customer service; it should be on completely rethinking it, and that requires the use of artificial intelligence (AI).\nCustomer service needs a rethink in the digital era\nCustomer service today is largely reactive. An individual calls into a business\u2019s contact center and needs to provide a wide range of information, such as name, address, loyalty number, contact information, and a description of the problem. If the agent can\u2019t help, the caller is passed on to the next person and the process starts all over again. There are many problems with this, most notably it wastes the customer\u2019s time and frustrates them.\u00a0\nImproving customer service isn\u2019t about making marginal improvements such as reducing hold times. Rather, it\u2019s about completely rethinking how customer service should work if enabled by AI. AI is able to connect the dots between a user\u2019s activity and information and make the agent seem almost like Kreskin, as they can predict why the person is calling and suggest how to solve their problem even before asked.\n\n[ Don't let these 5 things drag down your digital transformation; adopt the habits of highly effective digital transformations. | Get the latest on digital transformation by signing up for our CIO Leader newsletters. ]\n\nHow AI can help customer service\nTo illustrate, I\u2019ll provide a couple of examples of customer service with AI.\nAgent guidance without AI\nSarah Mitchell, a frequent traveler of Nationwide Airlines, is travelling from Denver to New York's JFK airport, but her flight is cancelled because of weather-related issues. The flight isn\u2019t until tomorrow, and she is given advanced notice from the airline.\nSarah logs onto the website and starts to search for new flights. As it turns out, all flights to JFK are either full or cancelled, so she expands her search to include LaGuardia and Newark in New Jersey. After spending 30 minutes, Sarah gets frustrated and calls the airline where she is greeted by a polite agent. The agent asks her name and frequent flier number and how he may help her. She explains the situation and the agent goes through many of the steps Sarah already went through. After a lengthy delay, Sarah starts to get agitated and is concerned the flights are filling up fast and she may miss out. Finally, after about 30 minutes, her flight is rebooked into a connecting flight to Long Island. It\u2019s not direct, like the other, but she is happy to get home, although irritated the transaction took so long.\nAgent guidance with AI\nAssume Sarah does all the same tasks up until the time she calls. When she calls the airline, the AI has gathered all the information and understands that her flight was canceled and that she had expanded her search to include other regional airports. The agent also knows that Sarah is a top-level flier and wants to ensure that she is re-accommodated as quickly as possible.\nWhen the agent answers, he greets Sarah by name and says, \u201cGood evening, Miss Mitchell, thank you for your loyalty to our Nationwide, I see you were trying to change your canceled flight. Can I help you with that?\u201d\nSarah is delighted the agent knew the information and says, \u201cYes, I am trying to get to the New York area, but my flight has been canceled.\u201d The AI has already figured out she is trying to fly to the closest airport to New York City and has found the connecting flight to Long Island. The agent says, \u201cI can put you on flight first thing in the morning to Long Island. Will that work for you?\u201d Sarah happily accepts it.\nIn the above example, based on her actions, the AI was able to understand who was calling, her status with the airline, and what she was trying to accomplish. And it provided an immediate response for the agent without requiring him to look up or ask for any additional information. This has the added benefit of getting the customer off the phone quickly, cutting down the hold time for other passengers that were affected.\nIntelligent Routing without AI\nDavid Thomas is a long-time subscriber to an internet-based music service and notices his bill is larger than he anticipated. He goes online to check his account and notices that he was charged for an album he purchased but then subsequently canceled.\nIn the search bar on the website, David enters \u201cErroneous billing\u201d and is given a number of options to choose from. He can\u2019t find what he is looking for and does some other searches. Finally, he gives up and calls the help number, where is greeted by an agent that asks how she may help him. He tells her the problem, and she states that she is sorry but she is in the online help department and can only answer questions about how the service works, and she transfers him to account support.\nDavid then needs to explain the problem to this agent, who informs David that account support doesn\u2019t handle refunds and transfers him yet again. After going through the explanation a third time, the refund is finally issued.\nIntelligent routing with AI\nWhen David calls the online music company, the AI has already analyzed the web search information and predicts that David is calling to ask for a refund on a charge that he feels is erroneous, and it routes the call directly to the refunds department. The AI pulls information from the portal, CRM system, and other data sources to analyze the customer history.\nCustomer service matters more than ever, and CIOs should look to AI as a game-changing technology. \u00a0\u00a0\nThe agent is quickly informed that David has been a customer in good standing for years and has never asked for a refund before. The agent greets David with, \u201cHello, Mr. Thomas. I see you feel you were erroneously billed. Can I help you with that?\u201d David explains that he purchased the album but then canceled it quickly. The agent informs him that accidents happen and the company would be happy to refund the money.\nIn this case, the AI was able to gather information based on David\u2019s activity and predict the purpose of the call. The results of the analysis were used to route the call to the appropriate person. The AI then executed business rules to determine whether the refund is approved and informed the agent.\nBusinesses are gathering massive amounts of information about the habits and activities of their customers. The challenge is that people can\u2019t analyze the large volumes of data as fast as machines. AI systems can be used to examine data and make inferences that can help businesses service customers faster and more accurately, making every customer service person more effective by putting the right information in front of them.\nCustomer service matters more than ever, and CIOs should look to AI as a game-changing technology.\n\nMore on AI and machine learning:\n\n A practical guide to machine learning in business \n 5 artificial intelligence trends that will dominate 2018 \n 9 machine learning myths \n Machine learning success stories: An inside look \nAI\u2019s biggest risk factor: Data gone wrong\n Winning the war for AI talent \n 9 IT projects primed for machine learning \n 10 signs you\u2019re ready for AI -- but might not succeed \n 10 strategic tips for getting started with machine learning \n Which deep learning network is best for you? \n How to build a highly effective AI team \n Why you should invest in AI talent now \n Why AI careers can start with a degree in linguistics \n The year of Alexa and the coming decade of A.I.