AG1 scaled from one SKU to nine in 18 months while maintaining 98-99% CSAT. Leala Francis reveals how she uses conversational AI to spot sentiment shifts before purchasing behavior changes, then adjusts supply forecasts mid-cycle. Her organizational design—consolidating customer happiness, insights, product commercialization, and packaging innovation under one role—creates the speed needed to launch four products without issues while keeping the feedback loop tight enough to catch problems when they're still "in the ones and twos."

The counter-intuitive insight: automation isn't about cost reduction at AG1. It's about relationship depth. By handling routine requests through AI agents that access full customer history and make personalized recommendations, human agents get freed up to send replacement shakers to customers they spot with old ones on social media. Leala's "birthday gift test"—would this interaction show we know the customer well enough to buy them the right gift?—defines whether personalization actually works at scale for a premium subscription brand expanding into retail.

Topics discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

Most enterprise CX leaders assume you need a CRM to manage customer relationships at scale. Aurelia Pollet, Director of Customer Experience at CarParts.com, shares a different perspective. Her team handles 43 million annual customers across 1 million SKUs while currently operating without traditional CRM infrastructure, and she explores how this approach might become more common as AI eliminates the integration complexity that made CRMs necessary.

The core insight: AI can potentially coordinate data across telephony, email, chat, SMS, order management, ERP, and accounting systems without requiring a central hub to force these disparate systems to communicate. This could address the "left hand not talking to right hand" problem that has challenged enterprise customer service operations for decades. Instead of building expensive integrations between systems, AI could act as the intelligent layer that accesses and synthesizes information from multiple sources in real-time.

Aurelia's implementation methodology centers on transcript analysis rather than theoretical workflow design. By examining actual customer service interactions across all channels, her team identifies which requests involve information provision versus complex problem-solving requiring empathy and advisory skills. This data-driven approach enabled them to deploy Spark, their AI assistant with access to the same data as human agents, while preserving human intervention for high-stakes scenarios requiring financial planning or technical troubleshooting expertise.

Topics Discussed: 

Listen to more episodes: 

Apple 

Spotify 

YouTube

What happens when a telecommunications reseller transforms into a cloud contact center pioneer? Paul Jarman, Founder and ex-CEO at NiceInContact navigated six major technology waves, from premise to cloud, single-tenant to multi-tenant, voice-only to omnichannel, and traditional analytics to AI-powered automation.

Paul has learned why most AI implementations fail the "looks good vs. gets used" test. His framework for AI adoption focused first on agent efficiency wins like automated after-call work, then real-time analytics, before attempting full automation.

But Paul's most contrarian insight centers on market consolidation. While everyone debates which layer will dominate — CRM giants like Salesforce, contact center platforms, or AI-native companies — he predicts CRMs will lead through acquisition rather than innovation. His reasoning: they have the market cap and megaphone, but lack the stomach to deploy thousands of developers for multi-year contact center rebuilds.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

What happens when you serve every single customer — from students to enterprise — with the same white-glove experience? At Notion, it created a CX challenge that traditional automation couldn't solve, forcing them to rethink the entire human-AI balance.

Emma Auscher, Global Head of CX, tells Ashish how Notion's "no customer left behind" philosophy led to a counterintuitive discovery: implementing AI increased both automated resolution rates and human agent interactions simultaneously. Rather than typical deflection strategies, they're creating what Emma calls "everyday luxury" experiences.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

Shantanu Das, General Manager & Global Head of Customer Service, Primary Sales, and Scam Prevention at Wayfair, manages one of the largest customer service operations in e-commerce — a 3,000-person global team at Wayfair — and his approach to AI implementation challenges conventional wisdom about automation versus human agents. His methodology starts with three distinct value creation modes: personal productivity improvement, real-time insight generation, and complete workflow reimagination. 

The breakthrough insight, he tells Ashish, came when he challenged his team to design systems as if no humans were involved, then strategically layered human expertise back into the process. This approach led to fundamental changes in how Wayfair delivers customer coaching, moving from monthly performance reviews to real-time feedback that happens in the moment of interaction.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

Managing millions of daily customer interactions at AT&T, Head of Retail Technology, Contact Center Platforms, GenAI Product & Engineering Deepak Sharma, has learned that successful AI transformation requires building AI-ready infrastructure before chasing AI features. His dual-lane framework separates quick wins like agent assist and call summarization from foundational data pipeline work that enables sophisticated AI at enterprise scale.

His most compelling example, he tells Ashish, involves digital avatars that create three-way interactions between customers, human agents, and AI, delivering experiences customers actually prefer over traditional service. Successful AI adoption happens when solutions are simple enough to feel like magic rather than technology requiring extensive training.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

The gap between AI promise and contact center reality is often measured in months of failed adoption and frustrated executives. Tyler Orrell, VP of Contact Center Operations at QuinStreet, tells Ashish how they developed a surgical approach to AI that focuses on business impact over technological sophistication. His framework for identifying the 6-7 behaviors that actually drive outcomes, rather than automating entire QA processes, offers a masterclass in strategic AI implementation. 

Tyler's contrarian vendor selection advice — never use the vendor's RFP form and resist "upper right quadrant" safe choices — challenges conventional procurement wisdom. His insight that insurance agents function as simultaneous consultants, salespeople, troubleshooters, and empathizers within single conversations explains why AI replacement timelines are more complex than most predictions suggest.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

As AI automation grows in customer experience, the most forward-thinking organizations aren't replacing humans, they're redefining how humans and AI work together. In this insightful conversation with David DeMarco, SVP of Business Technology at Carta, on AI CX Innovators, Ashish explores why increased automation actually makes quality assurance more crucial and how "white space mining" can uncover the 20% of issues driving 80% of CX improvements. 

David also shares Carta's strategic approach to channel selection, giving customers choice in how they engage while reserving human expertise for complex equity and valuation discussions. He also details their innovative AI workers program that's transforming coaching and sentiment analysis without complex rubrics—simply uploading a document with expectations generates comprehensive coaching plans across agent interactions. 

Topics Discussed:

When only 20-25% of customers complete satisfaction surveys — and even those are primarily negative experiences — how can you truly understand your entire customer base? In this episode of AI CX Innovators, Prasanna Chand, Head of Data & Digital Transformation at Wayfair, reveals how they're using AI to predict customer satisfaction scores with 85% correlation to actual survey results, providing a complete picture beyond the inherently skewed feedback pool.

Prasanna takes Ashish through Wayfair's journey implementing AI across their customer experience operations, from identifying critical issues within days of launching their loyalty program to helping agents self-coach through personalized insights rather than generic examples. With ChatGPT's launch as the tipping point, he explains how Wayfair strategically separated which AI solutions to build versus buy, and why their partnership with Level AI has been transformative for users across the organization.

Topics Discussed:

Listen to more episodes: 

Apple 

Spotify 

YouTube

AI isn't replacing humans in customer experience — it's transforming them. In our very first episode of AI CX Innovators, Ashish Nagar, Founder & CEO of Level AI, dives deep with inaugural guest Andy Yasutake, SVP and Global Head of Strategic Growth & Ventures at Edgevana. 

As former architect of customer experience transformations at tech giants eBay, LinkedIn, and Airbnb and with over 25 years shaping how global brands interact with millions of customers, Andy presents his battle-tested strategies for leading multi-million-dollar AI initiatives, navigating organizational resistance, and implementing generative AI at enterprise scale. 

From turning the 2020 pandemic into an opportunity for Airbnb's technology transformation to personally helping Brian Chesky deliver his vision of "11-star experiences," Andy shares candid insights few technology leaders have experienced across three waves of digital disruption.

Topics Discussed: