Game-Changing Coaching: What Sports Teaches Us About the Future of Workplace Development

Dr. Anna Tavis — Clinical Professor and Chair, Human Capital Management Program, NYU Steinhardt. Anna has written the book on the future of digital coaching, drawing on 15 years of executive experience in technology and financial services — including serving as a global head of talent development. Her research focuses on the intersection of coaching, technology, and performance, and she is a widely cited voice on why the skills-based talent model is due for an upgrade.

Parker Mitchell — Co-Founder and CEO, Valence. Parker leads Valence, the company behind Nadia, an AI coach deployed across dozens of Fortune 500 organizations to support leadership development at scale.

No one questions whether a 3-year-old at soccer camp deserves a coach. So why does the corporate world still need to prove that coaching works? Dr. Anna Tavis — NYU professor, former global talent leader, and author on digital coaching — joins Valence CEO Parker Mitchell to explore what professional sports already knows about coaching, feedback, and performance, and what that means for where AI coaching in the workplace is headed next. The conversation covers the limits of skills-based talent models, why psychological safety makes AI feedback more effective than human feedback in many situations, and what Olympic athletes can teach us about performing at the top of human ability.

Key Takeaways

• Sports already solved the coaching problem that organizations are still debating. In professional athletics, no one challenges whether coaching works or whether technology can enhance it. From youth soccer to the Olympics, coaching is simply assumed to be essential at every level. The corporate world is catching up — and the patterns established in sports provide a clear roadmap for what AI-enabled workplace coaching can become.
• Frequent, in-the-flow feedback is the single biggest driver of performance improvement. The failure of most corporate performance management systems comes down to one thing: the inability to provide ongoing feedback. Just as daily coaching feedback accelerates athletic performance, the equivalent in the workplace — timely, incremental feedback in the flow of work — is what will meaningfully move the needle on employee performance.
• AI feedback is psychologically safer than manager feedback in many situations. Research shows that employees feel significantly more comfortable receiving performance feedback from an AI system than from a manager. Status differences, availability constraints, and the corrective connotations of traditional coaching all suppress honest engagement. AI removes those barriers and creates a genuinely safe environment for reflection and growth.
• Skills are too one-dimensional to capture what performance actually requires. A skill applied in one environment produces a performance outcome; the same skill in a different environment may not. True performance depends on ambient context — team dynamics, psychological safety, situational awareness — that skills taxonomies cannot capture. AI is what will allow organizations to finally measure and develop performance in full rather than in fragments.
• The future of human work is operating at the top of human ability. Just as Olympic athletes stop consciously thinking about their skills during competition — relying instead on mental acuity, visualization, and judgment — AI will automate the delivery of many foundational skills in the workplace. What will remain, and what will require development, is the distinctly human capacity to navigate complexity, exercise judgment, and perform at the top of one's license.

Questions This Session Answers

What can professional sports teach us about the future of workplace coaching?

Professional sports — from Olympic competition to youth athletics — has already established that coaching is essential at every level of performance, and that technology enhances rather than replaces human coaches. Video feedback, performance data, and simulation tools have been integrated into athletic coaching for decades without controversy. The workplace is following the same trajectory: AI coaching tools are taking the role that sports technology has long played, delivering frequent, personalized feedback that no human coach could provide at scale.

Why does frequency of feedback matter so much for performance?

The core failure of most corporate performance management systems is the inability to deliver ongoing feedback — the shift from annual reviews to quarterly check-ins was a step in the right direction but still falls far short of what drives performance. Research on athletic coaching and workplace learning consistently shows that incremental, frequent feedback — ideally in the flow of work — produces meaningfully better outcomes than periodic formal reviews. AI coaching makes that frequency possible at scale for the first time.

Why is AI feedback sometimes more effective than feedback from a manager?

Research on AI-assisted feedback tools finds that employees feel significantly more psychologically safe receiving feedback from an AI system than from a manager. Several factors contribute: status differences between managers and direct reports create anxiety that suppresses honest engagement; managers are often unavailable or undertrained as coaches; and corporate coaching has historically carried a corrective stigma. AI removes those dynamics and provides objective, available, judgment-free feedback that employees are more willing to act on.

What is wrong with the current skills-based approach to talent development?

Skills are a necessary but insufficient foundation for performance. A skill applied in one context may produce excellent results; the same skill in a different team, under different conditions, may produce very different outcomes. Performance is shaped by ambient factors — psychological safety, team dynamics, situational context — that skills taxonomies cannot capture. Dr. Anna Tavis argues that AI will allow organizations to move beyond skills as the primary unit of talent measurement and begin to understand and develop performance in its full contextual richness.

What will human work look like as AI automates more foundational skills?

Olympic athletes do not think about their technical skills during competition — those have been automated through practice. They focus instead on mental acuity, visualization, and judgment under pressure. Dr. Tavis argues that the same shift is coming to the workplace: as AI takes over the delivery of many foundational skills, the distinctly human contribution will move toward operating at the top of one's ability — exercising judgment, navigating complexity, and bringing the best of human capability to situations that genuinely require it.

Full Session Transcript

You've written the book on digital coaching — why is the intersection of coaching and technology so important to you personally?

Parker: I'm going to welcome Anna to the stage. Anna is the chair of the Human Capital Management Program at NYU, and Anna has literally written the book on the future of coaching, the future of digital coaching. We're going to talk about some provocations around why coaching is so important, and what areas we can look to for a glimpse of what coaching in the workplace might look like. Welcome. Thank you, Anna.

Anna: Thank you so much. I want to build on — and maybe challenge a little — what we heard before about the future. Some of that future is already in the ecosystem and might not be sitting inside organizations. That's the topic I want to explore: where do we see coaching and technology already working together and setting precedent for what we could be looking for in our own organizations? What are the patterns that have already been tried, proven, and are working? That was the idea I had when researching my book, and I want to share it with you today.

Parker: Wonderful. Tell us a little bit about you and why the intersection of coaching and technology is so important to you.

Anna: I rebounded into academia. I was an academic, I left, and I spent 15 years in business — both in technology and financial services. That's where I realized how important this was. I worked in Europe, I worked here. As a head of global talent development, it always felt that coaching fell short as a tool, as a method. It was very effective with some people, and even with all of the investment we were making at the top of organizations, I don't think it was optimized — because it was a little too little, too late. The challenge has always been: how do we make coaching available and accessible at different levels in the organization?

The other thing is that most of the people in the audience know coaching was primarily applied as a corrective tool. Coaching got a bad reputation. If you told a senior executive on Wall Street that they needed to get a coach, that was a curse — like the next step would be a performance improvement plan. Coaching never got the impact it was intended to make. And when technology became available — first through platforms — I remember in 2019, as I was doing my research, there were whole conferences debating whether coaching would be effective on Skype. Or on the phone. Could coaching be done by phone? There was a lot of resistance. The desire to optimize what coaching could do — a method that goes back to the Greeks as historically the most effective form of learning — has been there for a long time.

Where do you look to get a sense of where coaching in the workplace is headed?

Parker: We know coaching works. We know that spreading it more widely — democratizing it — would be helpful. We're also looking at the future. So where do you look to get a sense of where workplace coaching is headed?

Anna: As I was looking around, sports — and specifically professional sports — was where I started doing research. It started with data. I remember "Moneyball." And when I looked at athletic performance, even the latest Olympics, there is no way those athletes could get to where they got without coaching. And it's interesting that we have this conversation here about trying to prove to our organizations that coaching works — because no one questions the effectiveness of coaching when it comes to sports. Even in little leagues, even at your basic soccer camp for 3-year-olds, no one is challenging the fact that all of those kids deserve to have a coach.

That barrier is already down in sports. You need a coach to get to any level of performance. So the question becomes: how do you get from basic training — where parents are doing the coaching — to even mid-level, school-level coaching? That's where technology started coming in a long time ago. From video feedback, to basic data points on the speed of your swing — getting those types of feedback to athletes early on doesn't mean you take away the human coach, the pro in your golf game. But technology was accepted from the start, as soon as those tools became available. And significant investment has gone into developing whole ecosystems of startups, with some clubs and professional associations building their own technology ecosystems to encourage innovation that accelerates coaching for professional athletes.

Parker: One of the words you used there was feedback. Technology can codify that feedback and deliver it at scale. Where does that analogy have a strong parallel with work — and where is work a little different from the sports world?

Anna: Everyone on this stage has talked about performance and productivity — and that's exactly where the Venn diagram sits between athletic performance and performance on the job. I've done a ton of research on performance management, and one of the main failures of current systems is the inability to provide ongoing feedback. If you look at an athlete and an employee side by side, it's the frequency of the conversation around feedback — just in time, in the flow of work — that makes the difference in the ultimate output of that employee.

Organizations started trying to solve this manually by mandating managers to have more frequent conversations. We moved from once a year to quarterly. There was a whole renaming revolution — check-ins and other types of language. But imagine if, just like with an athlete, you could provide feedback on almost a daily basis. There's no human capacity for that from the manager's side. The manager has to produce and give feedback — it was a real catch-22 in companies around performance management. If you had some way of providing daily, incremental feedback that builds up to the final performance appraisal, you'd have the same guarantee you have in sports: performance is going to improve.

Parker: Sports professionals can get feedback as they practice — they do a lot of practice before they actually perform. Managers and leaders, on the other hand, are constantly on the job. AI coaching can give them feedback in an extraordinarily safe practice environment, so they can build the confidence and skills to put it into practice in the real world.

Anna: I really want to emphasize that, because we've done research on using AI feedback tools. People feel a lot more comfortable and psychologically safe receiving feedback from an AI system than from a manager. There's a status difference with managers, and they're often not available and don't always have the capacity. It's a much safer situation to get objective feedback on how you're performing from an AI system.

Parker: There's a quick story I often share with CHROs. People say: the manager should just be the coach. But let me do a quick poll. How many people here have had a manager at some point in their career that they've had trouble working with? And how many here have someone on their team right now who's having trouble working with their manager? It's a bit of a trick question — but yes, we all have. We don't have a perfectly safe environment. AI coaching is, in many ways, far safer for a lot of people — it gives them a first draft of working through something before they bring it to a human relationship.

You're not a fan of where the skills conversation is headed — can you share why?

Parker: You and I have talked about a provocative idea around skills. The current path of skills-based talent models might not be the one you think is accurate. Can you share more about that?

Anna: I think we all agree that a skill is a very basic foundational building block of what performance really represents. And those of us working in organizations have heard a lot about context — psychological safety, team dynamics — and the many elements that contribute to ultimate performance. Skill doesn't guarantee performance. There are so many different elements that need to contribute to that perfect outcome. But our ability to measure the ambient context has always been very limited. Skills, yes — we can infer from how many words per minute you type, how fast you code. That's where it all started.

We're going to graduate from skills with the help of AI. We'll be able to contextualize performance — and maybe we need new language entirely, because "skills" has its own baggage from years of use. If we're able to have all of that wraparound context, in addition to identifying very specific granular skills, we're going to get much closer to actually identifying what it takes — if performance is our goal, which it is — to perform at the level of excellence required by a job.

Parker: That's such an important insight. Skills could be too one-dimensional. A skill in one environment might produce a performance outcome, and that same skill in a second environment might not. In a world where we can understand that context, that's a second dimension to the single dimension of skills — and it's so important to try to track.

Anna: I want to bring in the sports analogy again. All of those Olympic athletes we admire — they are not thinking about their skills when they're competing for gold or silver. They're thinking about mental acuity. Visualization. There are so many different elements they're focused on. The skill is automated by that point through years of practice. And I think we're going to see the same thing in the workplace. A lot of skills are going to be automated through AI and delivered to us. What will be required is a higher level: working at the top of your license, working at the top of your human ability — that is where we will need to compete.

Parker: I love that — the ability to move up that scale and bring the best parts of who we are, in the world of transformation we're all going to experience. Really appreciate the provocations for the audience. And thank you for joining us today, Anna.

Anna: Thank you.