The insurance labor shortage isn’t a hiring problem. It’s a budget one.

The demographic numbers get treated as a staffing crisis. They are really a budget one.

The National Association of Mutual Insurance Companies estimates 50% of insurance professionals will retire in the next decade.¹ The Bureau of Labor Statistics puts 28% of accountants and auditors over 55, with an estimated three-quarters of practicing CPAs having reached retirement age in 2020.² The US insurance sector is projected to lose around 400,000 workers by the early 2030s, with turnover in the sector already elevated from a historical 8-9% to somewhere between 12-15%.³

The response you hear from most brokerage principals is a mixture of concern and inevitability, with a plan that runs on hope. Recruit harder. Partner with community colleges. Build an apprenticeship program.⁴ All good ideas. All of them assume the right answer to fewer service staff is more service staff, hired at whatever the market now charges for them.

The core problem is that only 4% of millennials express any interest in an insurance career, and the industry has not made itself materially more attractive to Gen Z.⁵ Even if that number tripled tomorrow, the more useful question is not how you backfill every service seat that opens up. It is how many of those seats actually need to exist.

This is where the conversation has to change. The labor question is not how you keep service headcount at parity with the book as people retire out of it. It is how much of that headcount you can retire into software instead, and where the savings from doing that go. For most brokerages, the honest answer is producers.

Every brokerage runs on a set of processes that convert human hours into service, not new business. The commission accrual. The month-end close. The audit trail. The producer statements. The renewal analysis. Historically, the constraint on scaling any of this has been headcount on the service side: add work, add heads. What is changing is that AI can now do the parts of that work that used to require a person, which means the same level of service costs less to run. The question a principal actually has to answer is what happens to the difference.

The accounting profession has already lived through a version of this shift. Roughly 340,000 accountants and auditors have left their jobs in the last five years, a 17% decline, and 76% of US CFOs report a significant talent shortage in their finance and accounting teams.⁶ The firms that have adapted have done so by shifting what a human accountant spends time on, less reconciliation and preparation, more review and judgment, at a lower total headcount cost, not by hiring their way back to the old ratio.

The math is the same for a brokerage. If your close takes ten days and half of that time is spent assembling the numbers rather than deciding on them, the retirement of the person who assembles the numbers is not only a hiring problem. It is a chance to stop paying to assemble those numbers by hand at all, and then decide, deliberately, where that money goes next. For most agencies the honest answer is a producer, not a fourth account manager.

The reason we invested in Revori is that the AI story and the P&L story are the same story here. Every dollar you are not spending to backfill a service role, the reconciliation, the endorsement processing, the month-end close, is a dollar you can put behind someone whose job is to sell. The brokerages that come out ahead over the next decade will not be the ones running the leanest service teams for its own sake. They will be the ones that took the savings from a leaner service team and put it into production.

The labor shortage is here. The brokerages that treat it as a chance to rebalance spend toward production will grow. The ones that spend the savings backfilling the same service roles at the same ratio, or do not automate at all, will spend the next five years explaining a flat top line.