The Digital Executive

In this episode of The Digital Executive Podcast, host Brian Thomas sits down with Shenbo Xu, MIT PhD and co-founder of Kapnova, to make a distinction that most business leaders have heard but rarely internalized — and that costs companies millions when they get it wrong. Shenbo opens by tracing his path from MIT's AI lab through quantitative research at Point72 and foundational model training at Scale AI, explaining that each stop sharpened the same core insight: prediction tells you what will happen, but causal inference tells you what happens if you actually do something about it. Those are fundamentally different questions, and almost every AI system in deployment today is only built to answer the first one. He illustrates the danger with two vivid examples. First, a retail CEO who copied Apple's no-discount strategy without understanding that Apple's customers and his own were nothing alike — the sales crashed, and he lost his job. Second, eBay's discovery that years of spending on branded search ads — the ones triggered when someone literally types "eBay" into Google — produced almost no incremental value, because those customers were already going to the site anyway. The model said the ads were working. The experiment proved otherwise. This, Shenbo argues, is the gap that lives inside almost every ROAS number ever reported: prediction models get credit for outcomes that would have happened regardless. Causal AI builds an actual model of what drives what, so executives can ask what will happen if we do this before committing the budget. He closes with a forward-looking take on where causal reasoning is heading — pointing to recent DeepMind research suggesting that any AI capable of generalizing across new environments must have learned something causal — and argues that the next generation of decision-making tools will be built on mechanism, not just pattern, making them more honest, more durable, and genuinely trustworthy when the world changes.

If you liked what you heard today, please leave us a review - Apple or Spotify.

Learn more about your ad choices. Visit megaphone.fm/adchoices

In this episode of The Digital Executive Podcast, host Brian Thomas sits down with Katrina Purcell, growth operations consultant and former chief of staff, to explore the messy, often overlooked work that actually makes scaling companies succeed. Katrina opens with a career that wound through media, project management, and an executive MBA before landing her in the chief of staff role at IAS — where guiding a company through its IPO taught her that her real gift was bringing people along for the ride, not just building the strategy. She explains why she took that skill set independent: companies only IPO once, but dozens of founders need exactly this kind of help right now. She then tackles the most common failure mode she sees in leadership: keeping strategy behind closed doors until it is polished, then rolling it out to a team that never had a chance to shape it and therefore never truly owns it. Her prescription is consistent — involve people early, listen to the perspectives that complicate your plan, and recognize that your business has grown beyond just you the moment you hired a team. On goal-setting, she is equally direct: OKRs fail not because the framework is wrong but because companies copy what Amazon does today without remembering that Amazon did not start there, and because nobody owns the process or adjusts it as the company grows. She closes with a sharp observation about AI and operations — automation sounds appealing until you realize most teams have never documented their processes well enough for a machine to follow them. Her prediction is that AI will actually force the documentation discipline that most startups have always skipped, and that the founders who build durable companies will be the ones who invest in that foundation before they try to automate on top of it.

If you liked what you heard today, please leave us a review - Apple or Spotify.

Learn more about your ad choices. Visit megaphone.fm/adchoices

In this episode of The Digital Executive Podcast, host Brian Thomas sits down with Adam J. Schwartz, founder and CEO of DocumentReview.law, to explore what happens when a practicing attorney gets frustrated enough with existing tools to teach himself JavaScript and build a better one. Adam opens with the breaking point that sparked everything: a case with a few hundred documents where every existing platform was built for millions, making it both overkill and cost-prohibitive for the client. So he sat down in his converted garage in Los Angeles, dusted off his JavaScript knowledge, and spent two years building a prototype that eventually made professional developers chuckle — before they rebuilt it into a modern, browser-agnostic platform attorneys can use from any device. He then makes a clear-eyed case for where AI belongs in legal work and where it does not. AI can triage documents, surface the relevant ones faster, and eliminate the redundant early review passes that drive up billable hours — but until AI can stand in a courtroom, argue before a judge, or sit across from an opposing party in mediation, lawyers still need to know the case inside out. His governing principle is simple: trust but verify. He then pulls back the curtain on why document review consumes roughly seventy-five percent of discovery costs — the combination of hundreds of attorney hours and expensive vendor storage infrastructure — and explains how DocumentReview.law attacks both. He closes with a vision of legal practice where AI handles the ministerial work so attorneys can spend more time on the two things only humans can do: building the case and building the relationship with the client.

If you liked what you heard today, please leave us a review - Apple or Spotify.

Learn more about your ad choices. Visit megaphone.fm/adchoices

In this episode of The Digital Executive Podcast, host Brian Thomas sits down with Aidan McCarty, co-founder and co-CEO of Verified, to make the case that the way we prove who we are on the internet is about to look as dated as the fax machine — and that a phone number is the identity primitive that survives what comes next. Aidan opens with the Stripe analogy at the heart of Verified's thesis: Stripe didn't make payments better, it made payments disappear, and that is exactly what Verified is doing for signup. Where a typical fintech onboarding flow takes 161 seconds, 16 clicks, and loses forty percent of users along the way, Verified does it in 9.3 seconds — a user enters a phone number, confirms their first name, and their real identity fills itself in from authoritative sources. He then flips the intuition that speed and security trade off against each other: autofilling data from trusted sources is dramatically more secure than letting anyone type whatever they want into a form, because forms are exactly how fraudsters construct fake identities with dark web Social Security numbers. Regulators are catching up, with FinCEN, OCC, and FDIC jointly authorizing autofill as a compliant onboarding method. The conversation then turns to the bigger existential case: as AI agent traffic on the internet rounds up toward one hundred percent, the question of whether someone is actually a human — and which human — becomes the scarcest and most valuable signal on the web. A phone number, with its tenure, carrier data, activity history, and real-world cost, is almost impossible for an AI to fake and almost impossible for a human not to have, making it the missing identity primitive of the twenty-first century. Aidan closes with a prediction: forms will disappear the way fax machines did — not because someone built a better form, but because an obviously superior alternative already exists.

If you liked what you heard today, please leave us a review - Apple or Spotify.

Learn more about your ad choices. Visit megaphone.fm/adchoices

In this episode of The Digital Executive Podcast, host Brian Thomas sits down with Jen Grogono, founder and CEO of uStudio, to explore why the future of enterprise training and internal communications looks a lot more like Spotify than a company intranet. Jen opens with the founding story of ON Networks in 2006, where the challenge of syndicating original content across a hundred different endpoints — each with different metadata, formats, and technical requirements — planted the seed for what would become uStudio. When she saw enterprises building massive libraries of training and marketing content with no infrastructure to manage or measure them, the pivot was clear. She breaks down why the measurement layer is so critical to enterprise buyers: it is not enough to send a message, leaders need to know what is landing, who tuned out at the twenty percent mark, and what is actually changing behavior. That feedback loop, built on a secure instrumented platform inside corporate networks, is what separates content that moves people from content that gets ignored. She then addresses AI's role in two directions simultaneously — as the subject of urgent workforce change management that demands better communication, and as the engine making content creation, personalization, and discovery faster and more targeted than ever. Her product AutoCast turns corporate documents and knowledge bases into audio, so AI can surface the right three-second clip or five-minute deep dive exactly when a worker needs it. She closes with a vision of the enterprise as an always-on Spotify-like network — personalized to each worker's role, consumption habits, and learning needs, delivered through their phone at any idle moment — and makes the case that the companies that build that infrastructure now will be the ones whose messages actually land when it matters most.

If you liked what you heard today, please leave us a review - Apple or Spotify.

Learn more about your ad choices. Visit megaphone.fm/adchoices