For a long time, growth in insurance brokerage has come with a predictable consequence: more business creates more work.
More clients mean more submissions to prepare. More carrier requirements to navigate. More quotes or plans to compare. More proposals and client deliverables to build. More renewals to manage. More information moving between producers, account teams, clients, carriers, and systems.
And eventually, more people to keep everything moving.
That model is getting harder to sustain. Brokerages are facing talent constraints and margin pressure at the same time clients expect faster service and more strategic advice. Growth still matters, but firms can’t expect operational headcount to grow at the same pace.
I think AI creates an opportunity to change that equation.
The biggest opportunity isn’t simply to make brokers faster. It’s to help brokerages grow output and client value without operational complexity growing alongside them.
The problem isn’t any one manual task
Look at what happens when a commercial account goes to market.
The information needed to understand the risk may be spread across applications, loss runs, schedules, policies, emails, the AMS, and other documents.
Someone has to pull it together. Determine what’s missing. Prepare the submission. Navigate different carrier requirements. Track responses. Normalize quotes that come back in different formats. Identify meaningful differences. Then turn all of that into something the client can understand.
There are AI tools that can help with pieces of this.
Extract the loss run. Summarize a policy. Draft the submission. Compare two documents.
Useful? Absolutely.
But if someone still has to move the output of one task into the next, keep track of where everything stands, and coordinate the process from beginning to end, you’ve made individual tasks faster without fundamentally changing how the brokerage operates.
That distinction will matter more as AI becomes ubiquitous.
The real opportunity is to make complexity easier to absorb
Insurance brokerage isn’t a standardized business.
Different clients require different information. Different carriers have different appetites and submission requirements. Different lines of business have different workflows. And the process changes again when you move from new business to renewal.
Historically, that variability has made automation difficult. Standardize what you can, then rely on people to handle everything that doesn’t fit the template.
AI changes what’s possible because software can increasingly understand context, not just follow predetermined steps.
A brokerage can structure a risk once and use that information throughout the process. The workflow can identify what’s missing, adapt information to different requirements, help teams navigate markets, interpret quotes as they come back, and surface the differences that deserve a broker’s attention.
The goal isn’t to eliminate complexity.
It’s to stop making people manually carry that complexity through the organization.
Your best people shouldn’t be the integration layer
That may be the part of this shift I find most interesting.
In many brokerages, experienced employees are effectively the connective tissue between systems and processes.
They know where information lives. They know which spreadsheet needs to be updated. They know what a particular carrier expects. They know which details matter at renewal and which differences in a quote deserve attention.
That knowledge is incredibly valuable.
But using skilled people to repeatedly move, reconcile, reformat, and route information is an expensive way to operationalize it.
And it has another cost: every hour spent coordinating the process is an hour that isn’t spent with a client, navigating a market, advising on risk, or helping win the next piece of business.
AI creates the opportunity to put more of that operational knowledge into the way work itself happens.
The broker still owns the relationship, judgment, negotiation, and recommendation.
But the brokerage becomes less dependent on people remembering and manually executing every step required to get them there.
What this can look like in practice
This is the direction we’re building toward with Robin, Feathery’s AI operations assistant.
Instead of applying AI to one isolated broker task, teams can use Robin to build and adapt the workflow around the work itself, then interact with Robin as that work moves forward.
Scale should create leverage, not just more work
The specifics look different across insurance brokerage.
For P&C teams, the pressure may show up across submissions, carrier markets, quotes, proposals, and renewals.
For Employee Benefits teams, it may be census files, plan documents, renewal comparisons, benefit guides, and an expanding expectation that brokers provide strategic guidance beyond the renewal itself.
But the underlying challenge is similar.
How does the brokerage take on more business and deliver more value without requiring more operational effort at the same rate?
The traditional answer has largely been to add capacity around the complexity.
AI gives brokerages another option: redesign the operation so people spend less time coordinating work and more time applying the expertise clients actually value.
That matters for efficiency. But it also matters for growth.
A team that can respond faster, turn around client-ready work sooner, and spend more time advising clients has more capacity not only to service the existing book, but to win the next one.
Having AI won’t be the advantage
Soon, every brokerage will have access to AI.
Every AMS will have AI features. Every major software vendor will have an assistant. Producers and account teams will have copilots.
So access to AI won’t differentiate one brokerage from another.
What matters is what the brokerage changes because of it.
Does AI make a handful of existing tasks faster?
Or does it allow the firm to take on more business, deliver more value to clients, and adapt to change without operational complexity and headcount growing at the same rate?
I think the brokerages that figure out the second question will look very different from those that simply add AI to the software they already use.
The next generation of brokerages won’t just use AI to do the same work faster. They’ll use it to change how growth happens.
