New Qualtrics Tool will Simulate, Predict, Take Action

Qualtrics has trailed the launch of a new system called XM Data & AI, arriving in 2027. The new platform combines behavioral intelligence for customers, patients and employees, from two decades of Qualtrics work, with forty years of healthcare experience data from recently acquired Press Ganey.

Qualtrics' own data provides a continuously updating record across 18,000 organizations, while Press Ganey's spans some 41,000 facilities. The new system uses this to allow clients to simulate outcomes before committing to them; predict what a specific person is likely to do 'while there is still time to act', and 'deliver trusted outcomes, embedded inside the moment.'

The firm's focus is on continuity - it says the new platform 'replaces disconnected listening, survey applications and static experience data hubs.' It's introducing two buzzwords to underline this approach: while old-fashioned journey maps 'sit around, unused and covered in cobwebs,' Qualtrics says lifetime value is created across a continuous cycle of 'Experience Loops (X-loops),' not a single journey. These in turn carry the 'Experience Moments (X-moments)' that matter, compounding lifetime value.

The firm says is will use its understanding of experience to make this work: what a signal is, who it belongs to, where it sits in the lifecycle and what the next best outcome is. 'Without that layer,' it says, 'an AI produces answers that are plausible but often wrong. With it, Experience Agents produce trusted outcomes.'

To simulate and forecast the success of new launches, the platform generates synthetic data and builds digital twins of real customers. The company insists: 'These are not general-purpose models role-playing a persona. They are research-grade simulations, fine-tuned on Qualtrics' own research corpus, that you can use to pressure test price, policy, and product decisions.'

Meanwhile, the prediction capability promises to tell a client 'what a specific person is likely to do', using small language models called Experience Prediction Models, trained on Qualtrics' proprietary experience datasets and grounded on the firm's experience ontology. The third, Trusted Outcomes capability first checks that chosen actions chime with an organization's rules and policies, and only does what the client 'would have approved.' Only then is the recommended action taken by an Experience Agent.

Qualtrics - perhaps mindful of recent online discussion of its use of anonymised but proprietary client data to train more general models - stresses that in the new platform 'customer data never leaves your environment and never trains shared models... The advantage starts with your data, and it stays yours.'

Two new capabilities are available immediately: an X-Loop Diagnostic Service, which maps the customer lifecycle, identifies where data is missing, and surfaces the disconnected systems keeping the client from seeing what comes next, all with the help of Qualtrics XM Advisors; and a Reputation Experience tool, which brings customers' public comments into the XM suite, connected to the experience data that explains it.

Web site: www.qualtrics.com .