Why You Need to Meet Your Customers' Digital Twins
Artificial intelligence isn't just automating tasks, it's giving companies a virtual playground to see what would happen if they modify the customer experience without facing real-world consequences. That's thanks to a relatively new type of modeling that builds digital twins of customers.
With both physical and customer digital twins , companies can move beyond research conversations and run predictive experience simulations with expected real-world results. This expands the reach and effectiveness of CX programs, uncovers hidden financial opportunities, and enables operational experimentation companies might never try otherwise.
Through digitally learning how customers walk in their shoes or behave online today, company leaders can anticipate and act on tomorrow's operational needs. The result is a new frontier for CX, where virtual experiments directly shape real-world experiences in ways that were previously unimaginable.
What Are Digital Twins, Anyway?
A digital twin is a virtual representation, essentially a copy, of something, whether that's a physical place, an object, a segment of customers, or even a single individual. The idea is to take a copy of something that exists and use it to simulate and predict what would happen if you were to make a change to it.
The precursor to digital twins, physical twins, dates back to the 1960s when NASA built physical replicas of spacecraft to run simulations before sending live missions into space.
Thanks to the latest advances in AI, the concept is being reimagined for the 2020s. Digital twins of customers can be created using a combination of physical, interactional, emotional, psychographic, and other types of real-time data from across the customer experience, including customers’ actual behaviors, interests, attitudes, preferences, transactions, survey responses, and company interactions. They can be used in any scenario where there's value in learning from testing with a copy as many times as needed.
Take your typical grocery store or department within your favorite retailer. Thanks to years of research, there's a lot that's known about what impacts shopping behaviors, from how products are arranged and how store layouts are designed to how employees interact with customers. But there will always be new products to market and sell, new customer needs to meet, and even new customers to attract and serve.
With digital twins, retailers can go from following general best practices to unlocking specific insights into what changes to make for their in-store experience by observing actual and simulated virtual representations of the experience in action, whether that's moving ice cream from one location to another or altering the greeting employees use with customers, to increase average orders and repeat visits.
That's just what the home improvement retail giant Lowe's has done. It has taken virtual copies of its stores and used modeling, along with a combination of cameras, sensors, and data from customers interacting with its app while they're physically in-store to determine the pathways customers take through the store. The retailer has used these virtual representations to optimize how employees greet customers, offer help, and place products.
The Use Cases and Benefits of Digital Twins
Companies can use digital twins to simulate any experience before it's introduced in a real-world setting, enabling them to measure the impact of a new experience or a change to an existing one before going live. They can also be used to optimize every existing and new interaction across customer journeys and the overall lifecycle.
Companies have already realized value from early deployments of digital twins of customers. McKinseyhas observed revenue increases of up to 10 percent, on top of already realized benefits from physical twinning such as up to 90 percent improvement in speed of decision-making and up to 50 percent reduced development time.
While traditional market research and survey methods only allow companies to ask customers direct questions about their experiences, this AI-based approach allows organizations to identify customer friction, unmet needs, and gaps in the experience that might otherwise go undetected. It does this by connecting and giving a voice to data that's typically siloed and unanalyzed by experience professionals.
Digital twins and the associated what-if or predictive simulations can be used to improve nearly any experience across the business, including marketing personalization of mobile offers, conversational engagements across web, messaging, and voice channels, staffing and physical layouts, digital account signup and onboarding, pricing for new or enhanced products and services, cross-selling strategies, and agentic AI interaction resolution and cost management outcomes.
Companies can connect copies of their customers with their data lakes, meaning all of the data they have inside their organization, including finance, customer support, digital channels, marketing, IT, customer experience and employee experience programs and surveys, market research studies, and more. This creates ways for key team members such as strategists, data leaders, and decision-makers to test and validate their own hypotheses.
Questions like, "What would happen if we change this content?" and "What would happen if we move the breakfast bar in the hotel?" can be answered long before the final decision. And once decisions make their way to front-line operators, managers there can run their own what-if scenarios too, such as, "What would be the impact of rolling out this change without new training?" to control for localized customer or employee needs.
Experimentation and Innovation in a Safe Environment
It's important to keep in mind that, as with any AI technology deployment, model effectiveness and usability are only as strong as the readiness of the data, the tests being run, and the enablement to interpret and act on the results. These are all areas that should not be taken lightly.
That said, companies can and should try out digital twin and prediction scenarios, see the potential effects, and make adjustments as necessary in a controlled environment before going wide with a rollout or public initiative. That is exactly how great innovation begins.
The more data consumers share, the more convenient, seamless, and personalized they expect their experiences to be. At the same time, younger digital-native Gen Z consumers are entering the mix and want to find engaging and efficient experiences faster than ever.
Thanks to the rise of large language model-based and agentic interactions, companies are going to have fewer moments of direct connection with end consumers on their owned channels. That means fewer and shorter impressions to get things right, and using digital twins of the customer beforehand can help make the difference between winning or losing a consumer.
Industry leaders have discussed the potential of digital twins of customers since the early 2020s, but now, thanks to advancements in deep learning, generative and agentic AI, and integrated data lake and modeling solutions, we finally have the tools necessary to make what experts once only imagined a reality.
The future of customer experience will see companies increasingly delivering experiences that rely partially or fully on agentic interactions. AI virtual trials with digital twins of the customer will be crucial for stress-testing these experiences behind the scenes with the most realistic data possible before introducing them to real customers and employees.
Post-launch, digital twins can also be part of the continuous improvement loop to ensure refinement and experience improvement, especially before transforming workflows into something autonomous or adaptive.
Digital Twins and AI Simulations Will Shape the Future of Experience
Today's leaders are getting ahead by pushing boundaries and evolving their processes, tech stacks, and the experiences they offer. Now, thanks to digital twins of the customer and AI virtual trialing, they can do so with more precision, speed, understanding, and scale.
Customer experience teams with connected data sets and years of customer intelligence at their fingertips will be best positioned to use this powerful form of AI virtual trialing to predict the effects of CX changes, prevent friction, and uncover new operational opportunities. And when actioned as part of established frontline-ready AI programs, companies will realize deeper engagement, lower costs, and increased growth and loyalty among today's and tomorrow's customers.
Michael Mallett currently serves as vice president of digital go-to-market and strategy at Medallia, providers of an artificial intelligence-driven experience management platform that helps companies capture and analyze feedback signals from customers and employees to drive actionable business improvements.