This is part two of a two-part series on talent and organizational capability in insurance. Part one is here.
Imagine you read the first piece in this series and made a decision. You're going to hire differently. You're going to look beyond the familiar profile, find someone with genuine cross-boundary judgment, and put them in a role where that capability can matter.
What happens next is the subject of this piece.
In many insurance organizations, what happens next is this: The capability arrives and the organization fails to absorb it. Not intentionally. Not because anyone decided the new thinking wasn't valuable. But because the systems, processes, and approval structures that govern how work gets evaluated and decisions get made were built for something else entirely. The new hire learns, faster than anyone expected, how things work around here. And how things work around here tends to win.
This is not a hiring problem. It is an organizational conditions problem. And it is the reason that strategic ambitions stall in execution even in organizations with genuine commitment, real investment, and talented people who want to make change happen.
The operating infrastructure problem
Insurance organizations didn't build their operating infrastructure carelessly. They built it deliberately, over decades, in response to real requirements: regulatory compliance across multiple jurisdictions, shareholder and efficiency pressures, consistent underwriting standards, claims processes that could scale without introducing unacceptable variability. In the United States alone, an insurer of any significant size is navigating 50 separate regulatory environments, each with its own requirements, relationships, and oversight expectations. For global insurers, that complexity multiplies across every country in which they operate.
The operating infrastructure that manages that complexity is genuinely impressive. It is also, by design, optimized for consistency. Approval processes exist to catch exceptions. Review structures exist to ensure conformity with established standards. Escalation paths exist to bring unusual situations to people with the authority to make binding decisions.
All of that makes sense for the environment it was built to serve. The problem emerges when that same operating infrastructure is applied to the work of strategic change — which by its nature produces results that don't conform to established patterns, requires evaluation frameworks that don't yet exist, and depends on the willingness to learn from early signals rather than judge them against historical benchmarks.
The operating infrastructure isn't broken. It's optimized for the wrong conditions.
What this looks like in practice
I watched this happen firsthand at a large financial services organization. A capable digital team ran a test. The results were entirely normal for a first attempt in a new channel, with response rates that any experienced digital marketer would recognize as a reasonable baseline from which to learn and iterate. In the weekly meeting of the full senior executive team, a leader with significant authority over the program declared the results a failure — speaking with enough confidence that his view became the prevailing one in the room. With the CEO present and a culture that discouraged challenge, no one pushed back. A single comment, from someone with no digital or marketing background, applied with authority to work he lacked the context to evaluate, ended the conversation. The capability had arrived. The organizational conditions to use it correctly hadn't.
This pattern has three common expressions in insurance organizations today.
The first is evaluative mismatch: new kinds of work being assessed against criteria designed for old kinds of work. A digital initiative judged by the same ROI timeline as a mature product line. An AI pilot evaluated for consistency when the relevant question is what it's learning.
The second is authority distance: decision-making power sitting structurally removed from the market or operational reality being addressed. The people with approval authority don't have direct experience of the conditions the work is designed to respond to. Their judgment isn't wrong in the abstract. It's not calibrated for the specific context.
The third is what I'd call capability capture: the organization is so effective at socializing people into how things are done that genuinely different thinkers gradually stop thinking differently. It happens without anyone noticing. The cross-boundary judgment you hired for gets sanded down by the daily reality of operating inside a system that rewards conformity to established patterns.
None of these are personnel failures. They are system behaviors.
The data behind the pattern
BCG's research on insurance and AI tells a story that is easy to misread. Insurance, it turns out, is ahead of nearly every other sector in AI adoption. Insurers are experimenting, piloting, and investing. By the adoption measure, the sector looks like a leader.
The same research shows that only 7% of insurers have brought AI to scale across their organizations. Two-thirds remain in pilot mode.
That gap — high adoption, minimal scaling — is the organizational conditions problem made visible in numbers. It is not primarily a technology problem or a talent problem, though both matter. It is a problem of operating infrastructure. Pilots succeed in controlled conditions precisely because they operate outside the normal approval and evaluation processes. Scaling requires bringing the work inside those processes. And the processes weren't designed for it.
The question worth asking is not why so many initiatives fail to scale. The more useful question is what would have to be true about how the organization operates for scaling to be possible.
The capability gap underneath the execution gap
Closing the distance between ambition and execution requires building organizational capabilities that most insurance companies have not yet systematically developed.
The first is decision quality under uncertainty: the ability to make sound judgments when the data is early, the patterns are unfamiliar, and the right evaluative framework hasn't been established yet. This is different from risk management. Risk management is about known categories of uncertainty. This is about navigating genuinely novel conditions.
The second is coalition building across boundaries that don't naturally connect. Strategic change in insurance requires people in underwriting, technology, compliance, distribution, and customer experience to develop shared frameworks for evaluating work in progress. That doesn't happen through org chart alignment. It happens through deliberate capability building across functions that have historically operated in parallel.
The third is governance embedded in execution, as distinct from governance applied to execution. The difference matters. Governance applied to execution is a checkpoint — often a late one — where work is reviewed against compliance and consistency standards. Governance embedded in execution means the oversight function is integrated into how the work is designed and evaluated from the beginning, which allows for faster iteration, earlier identification of genuine risks, and fewer expensive course corrections.
These are not soft organizational development topics. They are the operating conditions that determine whether the talent you hire — whether traditionally credentialed or cross-boundary — can actually do the work the organization says it needs done.
The real question for insurance leaders
Insurance organizations are serious about strategic change. The investment is real. The intent is genuine. In many cases, the talent decisions are improving.
The harder question is whether the organizational conditions are keeping pace.
It is possible to hire for the AI era and still run a review process calibrated for a different one. It is possible to recruit cross-boundary judgment and then route every significant decision through an approval structure designed to enforce conformity. It is possible to announce a bold agenda and have it arrive at the middle of the organization as a set of directives that don't connect to how work actually gets done. A vision that generates genuine executive commitment can quietly lose momentum six months in — not because the commitment faded but because the organization lacked the operating conditions to carry it.
Nobody canceled it. It just stopped being anyone's job.
The gap between ambition and execution in insurance is not primarily a strategy problem or a technology problem. It is an organizational capability problem. And it will not be closed by the next hiring class, however strong they are, if the operating conditions and operating infrastructure they walk into were built for a different era.
The talent question and the organizational conditions question are not sequential. They have to be answered together.
