Positive customer experiences go beyond meeting or surpassing your customers’ expectations; they cultivate satisfaction and forge emotional connections that foster long-lasting loyalty and advocacy.

Introduction

Customer experience (CX) is a customer’s perception of your brand, products, or services. Their opinions are based on every touchpoint and interaction they’ve ever had with your brand - from first awareness to the most basic post-purchase service and support. Customer experience can be shaped by many factors, including the quality of your products and services, the effectiveness of customer support, the ease of navigation on digital platforms, the level of marketing personalization, and the consistency of interactions across various channels.

Positive customer experiences go beyond meeting or surpassing your customers’ expectations; they cultivate satisfaction and forge emotional connections that foster long-lasting loyalty and advocacy. Conversely, negative customer experiences can lead to dissatisfaction, customer churn, and irreparable damage to your brand reputation.

Organizations that prioritize and invest in top-notch data-led CX establish stronger customer relationships, gain a competitive edge, and nurture long-term customer loyalty, retention, and advocacy.

This investment is an investment in data. The type and amount of data needed for an effective CX strategy can vary widely depending on the size of the business, the industry, and the tactics employed. This data can include:

• Customer data. CRM data, onboarding data.
• Customer interactions. Customer interactions with website, mobile app, social media, email, and support, and customer feedback.
• Sales and purchase data. Customer transactions, purchase history, and order details.
• Demographic and psychographic data. Customer profiles, preferences, and behavioral data.
• Campaign metrics. Campaign performance, including click-throughs, conversions, and ROI.
• Market research data. Survey and focus group results and competitor analysis.

Bloomberg predicts the generative AI market to reach $1.3T within the decade1.

With each passing year, the volume and complexity of data have increased exponentially, presenting both a challenge and an opportunity for companies. And somewhat behind the scenes, the software to make sense of that data has also been maturing apace. Artificial intelligence (AI) — especially generative AI — makes it possible for machines to collect, analyze, and learn from massive amounts of data. So, while the sheer volume of data is a challenge for marketing, sales, and support teams to effectively manage, process, and extract valuable insights from, generative AI can quickly interpret, categorize, and translate massive amounts of data. It’s the perfect data-led CX partnership.

This white paper will cover how to apply generative AI to a data-led CX strategy, the benefits to be gleaned, AI’s challenges and risks, how we can meet those challenges responsibly and ethically, and how to keep pace with innovation. First, let’s get a basic understanding of AI and generative AI.

What are AI and generative AI?

At its core, AI makes it possible for machines to collect, analyze, and learn from external data. They learn by ingesting large amounts of unlabeled data and recognizing patterns in that data. Conventional AI’s primary uses have been limited to pattern detection, decision making, analytics, data classification, and fraud detection. But generative AI, the relative newcomer on the scene, has seemingly limitless potential use cases.

Gartner has been tracking generative AI on its Hype Cycle™ for Artificial Intelligence since 2020, but it was flying under the public radar until it made headlines in November 2022 with Open AI’s release of ChatGPT-3, a chatbot capable of very human-like interactions. This was not the first AI chatbot, but it was the first with such wide-ranging usefulness. Previous chatbots were primarily used by sales and marketing to guide potential customers toward a sale or help them with service issues. They had limited conversational abilities and could not create original content.

ChatGPT is a text-generating AI chatbot that understands context and responds with nuanced, contextually appropriate answers. It can remember previous interactions within the same conversation, answer questions, provide explanations, translate content, generate code, and write essays, poems, emails, and blog posts.

Within five days of its release, over one million people downloaded the free ChatGPT-3 app and began experimenting with its content creation capabilities. The critical outcome is that ChatGPT’s ease of use and broad utility encouraged professionals and the general public alike to explore GPT and other generative AI applications2. Soon, professionals in every sector were earmarking budgets for future use cases.

Gartner predicts generative AI’s impact will be “similar to that of the steam engine, electricity, and the internet.” AI is already in most homes, but before now, it was largely invisible: Siri, Alexa, and Google Assistant; Roomba; and the Nest thermostat are just a few examples of AI-embedded technology that don’t get the fanfare they deserve.

Since the launch of ChatGPT, everyone is talking about generative AI, but there were large language models (LLMs) in use long before OpenAI’s release of GPT. In fact, at EXL, we’ve been using LLMs in many of our solutions for years.

Generative AI is igniting a transformative revolution across industries, from enabling fully automated, personalized consumer marketing to generating concise medical reports for doctors. Its influence is being felt across the board, transforming the way we approach countless industries.

McKinsey calls 2023 generative AI’s breakout year. Adoption levels are high (especially in the technology industry), and in companies where AI is already in use, “40 percent expect to invest more in AI overall thanks to generative AI, and 28 percent say generative AI use is already on their board’s agenda.”

Transformation is underway. Disruption is imminent. But the risks are high.

Generative AI’s impact will be “similar to that of the steam engine, electricity, and the internet3.”

Generative AI is igniting a transformative revolution across industries, from enabling fully automated, personalized consumer marketing to generating concise medical reports for doctors.

How is generative AI transforming data-led CX?

Generative AI delivers the most impact in three business areas: marketing and sales, product and service development, and service operations (i.e., customer care, back-office support, etc.4) – incidentally, those are the areas comprising customer experience. And according to McKinsey research, when combined with software engineering, these areas deliver about 75 percent of the annual value from generative AI use cases.

Sales, marketing, service, and support can use generative AI as part of their data-led CX strategy in several ways, including.

• Personalization. An abundance of data provides comprehensive insights into individual preferences, behaviors, and interactions. Generative AI can leverage this data to create highly customized experiences for each customer, leading to improved engagement and satisfaction along the entire customer journey.
• Proactive customer experience. Approximately 30% of your customer service or contact center calls are follow-up calls (e.g., submitted claims, the status of a return, the status of an order, etc.). Generative AI can mine the data and push answers for those questions to the customer when the customer is most receptive via the contact center, app, text, email, etc.
• Predictive analytics. Businesses can employ generative AI algorithms to identify patterns, trends, and correlations to anticipate customer needs and proactively deliver tailored offers, recommendations, and content, and adapt product and service offerings.
• Real-time decision-making. Generative AI can quickly process and analyze vast amounts of data in real-time to enable data-driven decisions on the fly, ensuring swift responses to changing customer attitudes and market trends.
• Customer journey optimization. By combining data from various touchpoints throughout the customer journey, generative AI can help identify pain points and areas for improvement. You can then optimize the entire customer experience to enhance satisfaction and increase customer retention.
• Sentiment analysis and tracking. Generative AI can perform sentiment analysis on vast amounts of customer feedback, reviews, and social media interactions to gauge customer satisfaction, making it possible to address issues promptly.
• Market segmentation. Generative AI can identify micro-segments within a vast data pool, allowing sales and marketing to tailor their messaging and offerings more accurately.
• Advanced skills training. Generative AI can mine high-performing sales, customer service, claims, and customer support agent calls to identify the most effective tactics for each situation. This information can be used in training and as real-time prompts during live calls.
• Customer issue and claim resolution. Agent training covers about 90% of possible customer questions. For the other 10%, the agent can ask AI to search the thousands of knowledge base articles and return a solution, the steps to a resolution, and a proposed phone script — in seconds.

CX = 75% of annual generative AI value5.

Three CX use cases to start your generative AI journey

• Customer Engagement Bots move new prospects through their sales pipeline by providing contextualized information about your products while the prospect is actively researching.
• Digital Guides transform static Help and FAQ centers into conversational engagement by providing personalized responses to their questions.
• Knowledge Wikis reduce onboarding and response time for new agents by providing instant access to process maps, training guidelines, and knowledge bases.

The massive volumes of data available today present a wealth of opportunities to gain valuable insights, personalize interactions, predict customer behavior, and optimize the overall customer journey, leading to improved customer satisfaction, loyalty, and ultimately, business success.

But, as Winston Churchill once said, “Where there is great power, there is great responsibility.” As generative AI and other cutting-edge technologies drive innovation and progress, it is crucial for us to be mindful of the ethical implications and potential risks associated with their use.

What are the challenges and risks of generative AI?

One of generative AI’s biggest assets can also be one of its biggest risk factors — its depth of knowledge. To attain the depth of knowledge, large language models must be trained on massive datasets. For instance, OpenAI’s GPT-3 was trained on approximately 45 terabytes of text data comprising internet-based books, Reddit links, Wikipedia pages, and general internet pages – including its biases and inaccuracies. AI can amplify those biases in its decision-making and harm individuals.

Biases can emerge in several ways. For example, if an AI model is used in hiring, but the training data had gender or racial bias, the AI model may inadvertently perpetuate those biases when making candidate selection decisions. Or if a model designed to authorize bank loans or credit card applications was trained on data where certain groups are underrepresented, the model may lack sufficient exposure to accurately understand and represent those groups, leading to biased outcomes.

Ensuring fairness and mitigating bias in generative AI is not just a technical issue, it’s also an ethical one. As AI becomes more prevalent in applications, it is essential to implement rigorous evaluation and testing procedures to identify and address bias, making sure that AI systems are deployed responsibly and do not perpetuate harmful stereotypes or prejudices. As we harness the power of generative AI, it is essential to consider the impact on society, privacy, and the well-being of individuals.

Bias is not the only danger. There are other challenges and risks that need to be considered, addressed, and in some cases, regulated:

• Data privacy and security. Generative AI often requires vast amounts of data for training. This raises concerns about data privacy and the potential for data breaches if sensitive information is not adequately protected.
• Ethical use. There is a risk that generative AI can be misused for malicious purposes, such as creating fake content, deep fake videos, or spreading misinformation, which can have severe consequences for individuals and society. As generative AI becomes more sophisticated, it can be harder to distinguish between real and fake content, leading to potential misinformation and trust issues.
• Lack of transparency. Some advanced AI models, such as deep neural networks, can be difficult to interpret and explain. This lack of transparency may lead to concerns about the “black box” nature of the technology.
• Regulatory and legal challenges. The rapid advancement of generative AI may outpace regulatory frameworks, making it challenging to keep up with ethical and legal standards.
• Data overfitting. If not properly controlled, generative AI models can overfit to the training data, resulting in outputs that lack diversity and creativity.
• Human interaction and accountability. As AI systems become more autonomous, it is crucial to define human-AI interaction guidelines and establish accountability when things go wrong.

To address these challenges and mitigate risks, it is essential to implement responsible AI practices, ensure transparency and explainability of AI models, promote data privacy, establish ethical guidelines, and work collaboratively with policymakers to develop appropriate regulations. A comprehensive approach is necessary to fully realize the potential benefits of generative AI while safeguarding against its potential downsides.

45 terabytes of text data is equal to about one million feet of bookshelf space, a quarter quarter of the entire Library of Congress6, or about 3,748 million pages of text7.

To regulate or self-govern? Who is responsible for regulation?

To ensure the responsible use of generative AI and to address potential issues, we at EXL believe sensible regulation is crucial. This responsibility primarily lies with governments and regulatory bodies. However, as members of the business community, we should take the initiative to self-govern and work together to establish principles that can eventually become widely accepted regulations that benefit all parties. By doing so, we can foster a safer and more responsible environment for the use of generative AI technology.

At EXL, we advocate for two principles for the ethical, responsible use of generative AI:

• Train generative AI only on closed data sets to ensure the safety and confidentiality of all data.
• Ensure that the development and adoption of generative AI use cases have a “human in the loop.” (That is, make sure a real human checks the model’s output before it is published or used.) Avoid using generative AI models for critical decisions, such as those involving significant resources or human welfare.

We believe these principles are essential for maintaining accountability, transparency, and fairness. Responsible development and application of AI are paramount to ensure a positive and sustainable future for all.

We believe these principles are essential for maintaining accountability, transparency, and fairness.

How to stay ahead in a world of rocket-fueled innovation

When new technology is first introduced, there’s always a dilemma. Should your company be an early adopter and hope for competitive advantage and a say in future product enhancements or wait until the product is more stable before risking efficiencies and productivity on an untried solution? Facebook CEO Mark Zuckerberg probably said it best when he said, “The only strategy that is guaranteed to fail is not taking risks.”

With AI and generative AI, the speed of innovation may mean that more hesitation leaves your company in the dust. We have a better suggestion. Start moving. Even if you are only taking small steps. Move. Here are our tips for keeping pace during this lightning round of innovation.

Tip #1

Get leadership alignment

Build a business case that clearly articulates the opportunity AI can address and highlight the potential benefits, such as increased efficiency, cost savings, improved customer experience, or competitive advantage. Include key stakeholders from various departments in the discussions, gather their input, and address their concerns. Since you are planning a data-led CX initiative, why not start by putting together a Tiger Team of your product, marketing, sales, customer service, support, and software engineering leadership?

Tip #2

Embrace change

Take a proactive and adaptive approach to technology. Foster a culture of innovation and continuous learning within your organization. Welcome questions.

Tip #3

Prioritize data-led decisio n-making

Leveraging AI to gain insights from vast datasets will allow you to identify market trends, customer preferences, and potential areas for improvement. These insights will be stepping stones to more informed and strategic data-led customer experiences.

Tip #4

Pilot a project

Propose starting with a small pilot project that has clear business benefits – for instance, don’t deploy a chatbot if your business doesn’t have a clear business need for it. A good first project for a data-led CX initiative could be a real-time sales training chatbot based on the recorded sales calls from your top-performing sales reps. A successful pilot can build confidence among leadership and pave the way for broader adoption.

Tip #5

Be aware of AI comfort levels

Even as generative AI content becomes more human-like, there may be some resistance among customers and vendors to interacting with a non-human representative for claims, sales, or support. Be open and transparent about your use of AI and then supportive should an issue arise.

Tip #6

Focus on talent

Since its earliest history, artificial intelligence has sparked awe and fear in the hearts of man. They dream of creating a technology that is smarter than man (Done!), but they fear being put out of a job (Also done!). Encourage your employees to explore AI applications, attend training, and identify opportunities where AI can enhance business operations and further their careers. Give them opportunities to upskill.

Tip #7

Partner with an expert

EXL has more than 6,900 data scientists working across all aspects of AI, cloud development and data integration. More than 1,500 of these data professionals are dedicated generative AI experts who work with clients to drive deeper insights and advance business outcomes.
 

"At EXL, we help our clients understand the essential steps for successfully using generative AI, enabling them to extract real value from these solutions. Our expertise in core operations, including cloud-native infrastructure integration, dismantling data silos, predictive modeling, and identifying areas for workflow enhancements, means we possess a unique advantage in creating custom AI solutions using proprietary data sets that create superior AI value"

Rohit Kapoor
Chairman and Chief Executive Officer,
EXL
 

Conclusion

The transformative potential of generative AI for data-led CX is undeniable. The ever-increasing volume and complexity of data present both a challenge and an opportunity for companies. But generative AI offers a solution to efficiently process and interpret massive amounts of data, enabling businesses to gain valuable insights, personalize interactions, predict customer behavior, and optimize the overall customer journey.

The adoption of generative AI has already begun to reshape industries, from providing fully automated and personalized marketing to generating insightful medical reports and enhancing customer service. As McKinsey predicts, 2023 is poised to be generative AI’s breakout year, with significant investments and adoption across industries. However, it is essential for businesses to address the challenges and risks, including bias, data privacy, ethical use, and transparency. Responsible AI practices, collaboration with policymakers, and the adoption of ethical guidelines are crucial steps to ensure the safe and effective implementation of generative AI.

Engaging leadership, piloting small projects, prioritizing data-driven decisions, and fostering a culture of innovation are key steps to staying ahead in the AI environment. Partnering with experts like EXL, with our AI domain expertise, can provide businesses with the necessary support and cutting-edge solutions to leverage generative AI’s full potential. By embracing generative AI and data-led CX, companies can craft unparalleled customer experiences, build lasting customer loyalty, and create a brighter future for their organizations.

Reference

• https://www.bloomberg.com/ company/press/generative-ai-tobecome- a-1-3-trillion-market-by-2032- research-finds/
• IDC. “Generative AI Findings from Enterprise Intelligence Services Survey, 2023.” May 2023. Accessed August 4, 2023.
• https://www.gartner.com/en/topics/ generative-ai
• https://www.mckinsey.com/ capabilities/quantumblack/ourinsights/ the-state-of-ai-in-2023- generative-ais-breakout-year
• https://www.mckinsey.com/ capabilities/mckinsey-digital/ourinsights/ the-economic-potential-ofgenerative- ai-the-next-productivityfrontier
• https://www.mckinsey.com/featuredinsights/ mckinsey-explainers/what-isgenerative- ai
• https://service.alaska.edu/ TDClient/36/Portal/KB/ ArticleDet?ID=95

Written by:

Shashank Verma
Head of Data-led CX Practice

Ankush Jain
Vice President - Solution Owner - Data-led CX

Sanjay Pathak
Vice President - Data-led CX Capability and Delivery Head