Scaling GenAI from the Bottom Up

The Chief Talent Officer at Experian and former VP of People at Google cover the link between leadership and organizational performance, how to identify successful talent interventions, and how Experian is developing leaders at scale with AI coaching.

Table of Contents

5 min read

Going from use case to scale is the enterprise AI challenge of today. Most companies have started experimenting with generative AI, but only about 1 in 10 have actually found a way to deploy solutions at scale. 

Recently, we hosted a conversation between two HR leaders who have successfully scaled generative AI to build stronger leaders and higher-performing teams. Lesley Wilkinson, Experian’s Chief Talent Officer, is in the process of rolling out Valence’s AI-powered leadership coach as part of a leadership development initiative for all mid-level managers globally. Prasad Setty was Google’s former Vice President of People Operations. Today, he advises Fortune 500 HR leaders on their generative AI initiatives.  

Here are their most valuable takeaways for any HR leader trying to identify and scale an AI use case. 

Developing leaders drives ROI

[[simple-card]]

Teams that feel psychologically safe outperform others by 20% (as Prasad’s Project Aristotle at Google showed). In other words, a team’s performance is a direct reflection of its leader’s behavior. Good leadership interventions make leaders more effective, and more effective leaders improve organizational performance. 

Never stop experimenting

[[simple-card-2]]

With the proliferation of AI tools, the big challenge is figuring out which leadership interventions actually work. The best way to address this challenge? Give employees the ability to experiment with new tools.  

Failure is inevitable in the experimental phase of learning a new technology like GenAI, and the role of an HR leader is to facilitate and encourage experimentation, identify the most promising experiments, and then help scale those to the organization in a programmatic way. 

To do this at Experian, Lesley created a GenAI academy to encourage employees to experiment, discover new use cases, and share learnings. She then looked for what most excited and engaged leaders. Where engagement was high, she looked for a use case to scale. As she explained, “We look at an NPS for whatever we do. It's what outcomes we're driving. We try and keep that to the cleanest line possible. If we know leadership effectiveness is the cleanest route from an intervention to leadership effectiveness to organizational performance, that becomes our main measure.”

Scaling what works: Experian's AI coaching initiative

One of the tools Experian experimented with was Valence’s generative AI coach Nadia. Coming out of their first small pilot, Lesley had “never seen adoption statistics so high.” 

Once Nadia’s pilot succeeded, Experian focused on a strategy to tailor the AI to each individual manager. Leaders in the pilot found Nadia’s advice so helpful that they asked if they could use it to role play difficult or important conversations. Experian integrated internal datasets, such as 360-degree reviews, so managers could role play with guidance targeted to their specific context and individual strengths and weaknesses. 

The result? Effective leadership guidance and interventions at the critical moments that most directly impact organizational performance. 

Watch the full webinar below to gain more insights from Prasad and Leslie about scaling generative AI for personalized leadership development.

The following elements will be placed in between the rich text depending on where you set their attributes.The preview and this text won't be visible one the published site.
Attribute: [[button-cta]]
Attribute: [[divider]]
Attribute: [[big-cta]]

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique.

Placeholder Text
Attribute: [[simple-card]]

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

We have this beautiful straight line between the interventions that we select and leadership effectiveness, and leadership effectiveness and, at the end of that, organizational performance. That allows us to do a lot of experimentation, try out what we do, and measure, measure, measure.

Authors Name, Title

Lesley Wilkinson, Chief Talent Officer, Experian

Attribute: [[simple-card-2]]

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse 2 varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Really think about a whole bunch of curated use cases where you feel like you can get a lot of scale or a lot of value if it is deployed well. You are going to fail with some of those use cases, and that's okay. But unless you try out 10 or 20 things, you're not going to identify the three or four things that are going to be tremendously valuable for your organization.

Authors Name, Title

Prasad Setty. Former VP of People, Google

Attribute: [[simple-card-3]]

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse 3 varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.

Authors Name, Title

Attribute: [[quote-card]]

Ready to try it out?

Book a demo