The Engineering-First Incumbent: Why AI Transformation Requires Ownership, Not Just Software

Michael Lee of Sequence Holdings argues that the most significant value from AI will not come from startups disrupting incumbents, but from re-founding incumbents to leverage their existing scale, regulatory moats, and data density. The consequence of this model is that digital transformation as typically sold through consulting services or off-the-shelf software is incapable of producing deep systemic change. Because services firms optimize for billable hours and software vendors optimize for generic workflows, they avoid the radical reorganization required to truly integrate AI. For leaders, the advantage lies in recognizing that AI transformation is an organizational physics problem rather than a feature-deployment problem. Those who move from buying software to owning the operating system will capture the compounding value that others leave on the table.

The Hidden Cost of Fast Solutions

Most enterprises approach AI by purchasing software or hiring consultants to speed up existing human workflows. Lee argues this is a trap. Software vendors must design for the lowest common denominator of workflows to ensure broad compatibility. Consequently, they reinforce the current, inefficient human assembly line rather than replacing it with one optimized for autonomous agents.

> "What that means though, is that you are always going to sell to a workflow as it is designed today, as humans are set up in that human assembly line today. It is the only way you can sell a product. You cannot sell a product around a new human assembly line that does not exist today."
>
> -- Michael Lee

When you buy software, you are purchasing a static solution to a legacy problem. This creates a downstream effect where the organization remains tethered to its old operational logic, making it impossible to achieve the order-of-magnitude efficiency gains that AI enables.

Why Ownership is the Only Lever for Radical Change

The traditional private equity model is built on financial engineering and capital deployment. Sequence Holdings operates as a permanent holding company. This structural difference is important because it changes the time horizon and the incentive alignment.

When you are a vendor, you are incentivized to stay in the client wallet. When you are an owner, you are incentivized to reorganize the entire business to be AI-native. This requires a shift in celebrated persona. In most firms, the investor or the salesperson is the hero. Lee argues that if alpha comes from engineering, the firm must be culturally reorganized so that the engineer is the celebrated persona. Without this cultural shift, top-tier engineering talent will not stay, and the refounding will fail.

The 94% Efficiency Gain: A Case Study in Density

The most compelling evidence for this approach comes from Sequence investment in BankSouth. By treating the bank regulated nature as a feature, not a bug, and leveraging high data hygiene and centralized operations, they were able to re-engineer the underwriting process.

> "The average consumer underwriting today has decreased by 94% since we got started in March. ... Foundationally what we have done with the tools is that we have allowed the organization to be able to take on significantly more loan volume than was ever possible before with the same amount of headcount."
>
> -- Michael Lee

This is a classic systems-thinking result: by automating the rote, high-volume tasks, they did not just save money; they shifted the human capital toward high-value exceptions and field work. The system responded by doubling loan volume in a single quarter without increasing headcount or lowering standards. This is the lasting advantage of deep integration: the organization becomes capable of handling scale that would have previously broken the middle-office infrastructure.

Key Action Items

• Audit your celebrated persona: Over the next quarter, evaluate whether your organization prioritizes the salesperson, the investor, or the engineer. If your internal culture does not celebrate the people building the AI, your transformation efforts will be relegated to surface-level experiments.
• Identify dense operational nodes: Map your business to find where operations are centralized. These are your highest-leverage points for AI integration. Amortizing the cost of a custom agent across a centralized process creates a compounding return that fragmented, decentralized processes cannot match.
• Shift from buying to building workflows: Stop looking for off-the-shelf software to solve your biggest bottlenecks. If a process is unique to your competitive advantage, buying a generic tool will only lock in your inefficiency. Invest in internal data ontology, making your business legible to models, over the next 12 to 18 months.
• Reframe change management as human engineering: Recognize that the anxiety caused by AI is a signal of impending structural change. Prioritize transparency and role-redefinition, moving staff from rote processing to high-value, human-centric roles, to convert organizational resistance into operational momentum.
• Adopt a one-deal focus: For leaders managing multiple initiatives, prioritize depth over breadth. Attempting to transform five processes at once often leads to shallow, ineffective results. Focus on one core workflow until you achieve the 94% efficiency type of breakthrough, then use that as the foundation for the next.