Could you share a bit about your background and what led you to found Frame AI?
I’ve been working in the AI space for the entirety of my 20-year career. During my Ph.D. at Carnegie Mellon University, I published research in game theory, AI, and multi-agent systems. Through that time, I’ve become focused on the practical use cases for leveraging AI. I’ve primarily worked on projects that utilize AI technology as a means by which to help humans make strategic decisions with more data than they consistently gather on their own.
What is the overarching vision of Frame AI? How do you envision the impact your platform will have on the customer experience landscape?
Frame AI provides the framework for companies to measure and understand customer feedback, completely independent of surveys or reliance on the human eye to catch and contextualize insights from recurring themes captured across the spectrum of CX tools. Frame AI’s CX scores provide simple, actionable leading indicators essential for staying ahead of customer expectations and protecting revenue.
Ultimately, our vision is to unlock the value of the massive, dormant datasets companies have amassed over the last decade or so. So far, much of that data sits either completely unused or drawn very occasionally into manual analysis. With LLMs and machine learning, we can automate proactive analysis, sourcing insights that would otherwise go completely unnoticed.
In the context of AI-driven customer insights, how do you balance the need for automation with the importance of human interpretation and decision-making?
Right now, most of the AI products being marketed are automation – tools that replace human work, often at a fractional cost but also with quality tradeoffs. A few categories, such as agent assist, enterprise search, and writing assistants, deserve the title augmentation: they help a human perform their job exceptionally, rather than replacing them.
We want Frame AI to go one step further, to what we call orchestration: helping organizations collaborate better. Orchestration happens when you proactively bring data and insights that individuals didn’t know to ask for, sourced from across the organization and delivered to where they work within the organization. With orchestration, each team’s efforts enable all others, not by replacing their work, but by proactively helping them make better decisions.
How does Frame AI utilize AI to identify pain points in customer interactions? Could you walk us through the process of how AI analyzes customer interactions to uncover these pain points?
A very simple way to think about it is that our AI platform sifts through all available customer experience channels – chats, agent call recordings, support tickets, email, etc. – to find recurring themes. Suppose customers consistently return a specific pair of jeans with a specific skew number, and they share a specific reason, such as “poor materials” in their reason for returning said jeans. In that case, the AI can identify that a specific skew is seeing a recurring mention of “poor materials” from customers, regardless of whether they call, chat, or email their concerns.
Coming to such a conclusion across the entire spectrum of a business through the eyes of a human would be inefficient at best and ineffective at worst. We use AI to do the heavy lifting so CX leaders and executives can focus on the high-level insights rather than figuring out how to source the insights in the first place.
In what ways can businesses utilize the insights gained from identifying pain points to enhance customer experience and increase customer lifetime value (LTV)?
In the ideal scenario, using Frame AI will allow customer experience teams to move from being a cost center to a revenue protector (and expander). Over time, quickly identifying, addressing, and then preemptively remedying customer challenges will (1) preserve the money customers have already spent with you and (2) make them more likely to spend money with you in the future.
What measures does Frame AI take to ensure that the insights extracted from customer interactions lead to concrete revenue-related actions?
Frame AI boils down the details of AI enrichment into what we call “CX scores,” which provide simple, high-frequency leading indicators with an average dollar amount tagged to the issue at hand. To continue on with the pair of jeans made of “poor materials,” a CX leader and their executive will see the price of the jeans themselves, restocking, shipping, and other associated costs to present – on average – the exact amount in lost revenue tied to each instance of this recurring issue.
What security and privacy measures does Frame AI implement to ensure the protection of sensitive customer data?
While the popularity of AI has waxed and waned during my career, awareness of the value of data has only gone in one direction: UP. In the 2000’s, that meant people building data marketplaces. In the 2010’s, it meant enterprises seeking to exercise more control of their own data through improved security. And in the 2020’s it means a global effort to preserve individual rights over their own data through GDPR, CCPA and other government regulations.
We’ve had the luxury of building Frame AI with full awareness of this trend, and with the benefit of cutting edge work on security and privacy practices. Since the early days, we’ve leaned into standards for these practices, maintaining audited SOC II Type 2 certification and audited HIPAA compliance programs.
We’ve also learned by working with companies in sensitive and highly regulated industries such as healthcare and financial services. While many businesses use Frame AI as a SaaS platform, a key to serving these industries is being able to deploy Frame AI completely within a customers’ data environment, so that they maintain total control over data going in and out. We can do this because our own business doesn’t depend on mixing data across clients – each deployment stands on its own.
Marketing data often exists in silos across various platforms. How does Frame AI help in bringing coherence to this data?
This is one of the top promises of Frame AI. Over the last 20 years or so, marketers, in particular, have become adept at housing massive amounts of data about each one of their customers. Though the methods of sourcing and storing these data sets are many, making sense of it all, separate or together, is often just short of impossible.
Frame AI can look across silos, including data lakes, CRMs, and CDPs, to search for and then surface insights that no individual human is in a position to identify.
When we talk about “coherence”, I’m not necessarily referring to a single view of the customer, or “Golden Profile”, to serve every application. We believe it’s important for insights to be tailored for relevance to specific teams in order to be actionable. What a compliance manager, marketer, or contact center leader needs to know about the same customer are different views! But by basing each of them on a common set of enriched data, we can make sure that each team’s actions are aligned, and that the customer experience is itself coherent.
In terms of enhancing marketing effectiveness, can you highlight some key advantages that result from having coherent and consolidated marketing data?
To stand out in a crowded attention landscape, marketers must provide the most personalized experiences they possibly can. And increasingly, they must do so primarily with first-party data.
For the marketers, as with other CX leaders, this allows them to meet an expectation that is incredibly reasonable on the part of consumers, but almost never achieved: that, if I share something with a brand on one channel, they should now “know” this about me and act accordingly. We think reaching this goal will help marketers drive both better campaign success and better brand-consumer relationships overall.
For more such updates, follow us on Google News Martech News
George Davis, Founder and Chief Executive Officer at Frame AI
George Davis, founder and CEO of customer experience intelligence, AI platform, Frame AI. George holds a Ph.D. from Carnegie Mellon University and has published research in game theory, AI, and multi-agent systems. LinkedIn.