Insight & Analysis | Technology & Business Solutions (TABS)

The AI Copilot Culture

July 25, 2025

Key Takeaways:

An emerging trend in AI adoption is the rise of the “copilot culture,” where employees across roles are empowered to use AI tools that enhance their individual workflows.
During THL’s Gen AI Innovation Day, portfolio company leaders shared how they are building a culture of AI adoption.
The bottom line? AI is only as powerful as the people and infrastructure behind it—when companies invest with intention and invite employees to shape how AI gets used.

There’s no denying AI’s potential. But even the most advanced tools aren’t going to drive impact if they’re not used effectively—or at all. For companies to succeed with AI, employees need to feel empowered to not just use it, but to explore how it can support their unique workflows and work styles. What helps on person may not help another, even if they’re on the same team or using the same systems.

Effective AI strategies require a culture shift—one that is already underway across THL’s portfolio. In each of the firm’s three core verticals—Technology & Business Solutions, Healthcare, and Financial Technology & Services—teams are embracing what it means to have a copilot culture: a workplace where employees are equipped and encouraged to integrate AI into their workflows. And that goes for all employees—not just engineers and tech leaders.

At THL’s recent GenAI Innovation Day, presentations from portfolio companies Bazaarvoice, Centria, and Hexure illustrated this shift in action and inspired attendees to think about their own AI cultures.

“GenAI is in everyone’s life, but it’s still early innings,” said Mark Benaquista, Managing Director, THL. “How do we take advantage of that? How do we learn from each other and keep momentum going? That’s what our Innovation Day was for—and it’s clear that so much of AI’s potential comes from building personalized solutions that are deeply rooted in the way people work.”

Bazaarvoice: Empowering Every Innovator

Bazaarvoice: an e-commerce enablement software company whose full-funnel shopper engagement platform powers the collection, curation and display of user-generated content like ratings and reviews, photos, videos and Q&A across the digital shopping journey.

Lou Kratz, Senior Principal Engineer at Bazaarvoice, began his presentation with a bold – deliberately provocative – proposition: “AI sucks.”

Clarifying that the technology’s hype outpaces the everyday reality for its users, Kratz acknowledged that, even with AI’s power and potential, it adds complexity that frustrates users. Giving a personal example, Kratz spoke of a grocery delivery app that recommended jelly simply because he added peanut butter to his cart. The app ignored his intent—meal planning for the week—and instead focused on obvious pairings.

Similarly, professional GenAI tools sometimes offer centralized one-size-fits-all solutions that sound great on paper, but don’t meet the nuanced needs and workstyles of individual employees.

No wonder Kratz has made it his mission to bridge the gap between AI capabilities and real user needs. His team is pioneering a bottom-up approach, empowering non-technical employees to build their own lightweight copilots using no-code tools like Dify:

AI Chatbot for Client Success: Bazaarvoice’s SVP of Client Success flagged that customer success managers were wasting four hours a week searching for internal information across fragmented systems. Kratz’s team built an Amazon Q chatbot that cut search time in half—unlocking $250,000 in productivity.
Battle Card Generator: The company’s Director of Revenue Enablement created a prompt in Gemini to help sales reps generate competitive battle cards on demand, but the manual copy/paste process and outdated data made it hard to scale. So Kratz rebuilt the tool using Dify, turning it into a live app that scrapes real-time data, applies the original prompt, and produces sharable, up-to-date battle cards in under two minutes.

“Every single person has unique needs where AI could help, but the generic tools are falling short,” said Kratz. “What if unlocking 10x growth isn’t about deploying different AI tools but empowering every single employee to wield AI to solve their unique problems?”

After speaking with hundreds of Bazaarvoice employees to understand their pain points, Kratz has a backlog of 87 employee ideas for how AI can make their job better. He’s challenged himself to build solutions for them all this summer.

Centria: AI to Augment, Not Overhaul

Centria: is a leading provider of Applied Behavioral Analysis services to children diagnosed with autism spectrum disorder.

At Centria, AI is not just the domain of engineers—it’s a practical tool being placed directly into the hands of frontline teams, recruiters, and support staff. Copilots are quietly transforming some of the most complex and operationally heavy corners of the business.

A Teams-integrated chatbot called Scout handles over 100 employee inquiries a day—from PTO balances to payroll troubleshooting—reducing ticket volume by more than 25%. PayerGPT, another specialized chatbot, simplifies complex payer contracts and manuals, which often span hundreds of pages and vary by state and code. A Java-migration assistant uses generative AI to accelerate legacy code conversion—and cut the cost of a two-year project down by 67%. Meanwhile, the company’s voice AI interview tool automates high-volume recruiting by calling applicants within 30 seconds of applying, conducting dynamic interviews, and summarizing results for fast decision-making. The result: the company is able to hire and train more technicians in order to better serve more families in need of ABA therapy.

Clearly, Chief Information Officer Pawanjit Singh doesn’t want to build sweeping, monolithic AI systems—instead, he wants to embed task-specific copilots where they’re needed most.

“You just have to deconstruct the problem down to the simplest thing,” he said. “When you do that, then you can create certain AI agents that can become an augmentation.”

Learn more about Centria’s agentic AI strategy—and the broader promise of agents across industries—in our companion whitepaper, “Agentic AI at Scale.”

Hexure: Turning Hackathons into Habits

Hexure: A software company whose core product streamline sales for life insurance and annuity products by connecting insurance product originators with distribution partners.

When Warren Perlman joined Hexure as CTO, he inherited legacy codebases and a development culture ready for reinvention. His three-pronged approach—AI to build product, AI to supplement product, and AI in the product—set the tone for a transformation grounded in enablement and efficiency.

“GitHub Copilot is the choice that I made for our organization, and what it allowed us to do was really accelerate delivery, but that wasn’t my problem,” he said. “My problem was getting people to actually use it.”

He thought a hackathon might help spur adoption, so he encouraged anyone on his 90-person team to use Copilot to build anything, whether it was related to their core product or not.

“I told them to pick an idea, make it work, and then come back and tell my why it worked,” he added. “About 30 people participated and those 30 people became the people who led the organization from an AI perspective.”

Their success proved Copilot’s value and helped turn initial skeptics into adopters. Now, the majority of his organization is using Copilot, developers onboard in weeks instead of months, and efficiency is up 45% in development and 65% in sprint planning.

The building of AI goes beyond the hackathon. The company continues to encourage people to build solutions with AI and experiment with AI in the Azure realm.

Looking Ahead

These presentations illustrate what’s possible when companies embrace AI as part of their culture—equipping employees with no-code tools and encouraging experimentation across every department and at every level. With a copilot culture, employees don’t just adopt AI—they help guide its implementation.

“At THL, we believe gen AI’s greatest potential lies in impowering employees to shape the tools they use every day,” said Jagjit Singh, Director, THL. “That’s what creates a flywheel effect—where ideas become solutions, tests become the norm, and results become proof points that inspire broader adoption.”

To learn more about how THL and its portfolio companies are embracing a copilot culture, visit www.THL.com or contact our team today.

About GenAI Innovation Day 2025

In June 2025, THL—together with AWS—hosted its first-ever GenAI Innovation Day at AWS’s Boston office, bringing together nearly 50 product, technology, and AI leaders from across the portfolio. The day-long event was part of THL’s broader push to foster collaboration and knowledge-sharing around AI implementation. It serves as a hands-on complement to the firm’s quarterly AI roundtables. To learn about other topics discussed at Innovation Day, read “Agentic AI and Scale” and “What It Takes to Scale AI.”

—

Mark Benaquista, Managing Director, SRG
mbenaquista@thl.com

Jagjit Singh, Director, SRG
jsingh@thl.com

Alex Sabel, Vice President
asabel@thl.com