A company’s history can often be split into two distinct acts: the first decade spent disrupting the market, and the second spent desperately fending off disruption from others. That’s the pattern Raghu Raghuram observed leading VMware, a cycle now playing out in real-time as giants like Cisco navigate the seismic shift brought by AI. The conversation delves into what it truly takes to steer a massive organization through technological earthquakes, moving beyond simple innovation to a fundamental rewiring of mindset, structure, and story.

The core of the discussion revolves around how large, successful companies inevitably lose touch with the “front lines” where real disruption begins. Leaders get trapped talking only to their biggest enterprise customers, while the next wave—be it cloud, containers, or AI—emerges from entirely different users, like developers or individual employees with credit cards. To combat this, both Raghuram and Cisco’s Jitu Patel emphasize the necessity of operating with “founder-like urgency,” even at scale. This means creating small, ring-fenced teams with direct air cover from the top to build new products, and having the discipline to define a narrow “ideal practitioner profile” for initial adoption, rather than trying to sell an unproven product to the largest, most demanding clients immediately.

A successful transformation, they argue, must be product-led and story-driven. Patel stresses that “the story is the strategy,” and leaders cannot delegate storytelling in a large company; one clear, compelling narrative is essential to align tens of thousands of employees. Furthermore, they dissect the nuanced approaches to innovation—knowing when to build organically versus acquire, and understanding that to displace an incumbent, your new solution must be an order of magnitude (10x) better, not just marginally improved. As the AI wave accelerates, they see it fundamentally breaking old sales models and reshaping infrastructure demands, from networking and power to security, presenting both an existential threat and a massive opportunity for those who can adapt.

Surprising Insights

• Large companies are great at running experiments but terrible at doubling down on the ones that work. Startups excel at focusing on one problem; big companies often lack the muscle to aggressively scale a successful experiment into a new business.
• To displace a mature competitor, you must first be prepared to coexist with them. A common mistake is trying to build a complete, greenfield replacement. Winning requires identifying an insertion point in the existing (brownfield) market and coexisting before you can systematically displace the incumbent.
• The ideal customer profile for a new product should focus on the “practitioner,” not the “company.” Successful internal incubations nail the profile of the individual who will use the product daily, then expand from that beachhead. Starting by targeting a massive company like J.P. Morgan for a version-one product is a recipe for rejection.
• In the AI era, you may need to unapologetically partner with your largest competitors. If a competitor owns more than 20% of a market, not integrating with them only excludes you from that market. Cisco cited partnering with Microsoft Teams, a direct competitor, as a key driver of hundreds of millions in revenue.
• “The story is the strategy.” For a company of 95,000 people, a set of bullet points is useless. A simple, clear, and well-told story is what galvanizes action and aligns the entire organization; it’s the most practical form of strategy.

Practical Takeaways

• Ring-fence innovation teams and give them direct air cover. Start with a “two-pizza team” (small enough to feed with two pizzas) that is protected from corporate antibodies and has agency. Their two jobs are to build a great product and define the initial, narrow ideal practitioner profile.
• Leaders must own the narrative. Do not delegate the company’s core story to marketing or communications. The top leader must be the primary storyteller to ensure one clear voice and prevent strategic fracture across a large organization.
• Seek the truth with binary language. Combat large-company politeness by having brutally direct conversations. Use clear, binary terms like “we are failing in this area” to outline problems, and follow with the specific, critical actions needed to succeed.
• Design reviews should be title-less and safe. To get to the best ideas, create forums where designers, PMs, and engineers can critique work with titles left outside the room. The goal is to make it a safe space where the best idea wins, not the highest rank.
• When building a new product in an existing category, aim for a 10x improvement. Do not play catch-up. To extract an entrenched incumbent, you need an order-of-magnitude better solution, whether in cost, performance, or capability. Be the biggest skeptic of your own claims to ensure you’re not settling for 15%.