Where should MGAs leverage AI today?

Managing General Agents (MGAs) hold an advantage over traditional insurers, filling gaps in coverage and offering the agility and innovation the market requires. The demands are significant: achieve more with less, maintain strong broker and capacity relationships, and keep the submission-to-bind process efficient.

Yet technology, and AI in particular, has not always delivered on expectations. According to an MIT report from July, enterprises have invested $30–40 billion in GenAI, but 95% report zero return. Executives are cautious. They look for trusted vendors, tools that integrate with existing workflows, clear boundaries on data use, and systems that improve with time. What they often experience instead are fragmented tools and pilots that fail to yield results.

For MGAs, value comes from AI applied to real pressure points. With the right foundation, it is possible to speed up submission-to-quote, refine pricing models more quickly, protect rate adequacy at the point of pricing, and capture data in a way that strengthens broker and capacity conversations.

What Should a Great Technology Stack Help an MGA Achieve?

The critical question is whether an MGA’s technology stack strengthens agility and credibility, or whether it is constrained by legacy systems and outdated practices. Spreadsheets, disconnected tools, and fragmented data limit the ability to build a portfolio asset, review historical performance, and generate reporting that highlights opportunities.

Artificial intelligence and automation have become central to an MGA’s ability to:

1. accelerate submission-to-quote turnaround for brokers and partners

2. deploy and refine pricing models at speed

3. protect rate adequacy at the point of pricing

4. capture insights that help adapt product design to emerging risks

The sections below explore each area, showing how MGAs can address the challenges with data and AI.

Quote Turnaround

How many submissions can an MGA review? From our engagements, unproductive firms handle between one in five and one in two. If only 20% of those are underwritten, that leaves 4–10% of opportunities bound. Brokers experience delays, declined submissions, or no response.

The missed opportunities are clear, but the hidden cost is the lost insight. Submissions at scale can inform reporting, recalibrate underwriting guidelines, and highlight product gaps.

AI-driven ingestion and cleansing changes this picture. Instead of underwriters working line by line through statements of values, platforms like hx Renew automate triage, enrich data, and free up underwriters for risk evaluation. Throughput improves, and brokers receive faster responses. In practice, this is where Ingestion Agent adds value: automating intake, cleansing, and schema mapping while underwriters remain in control. Hours are released back to underwriting teams for higher-value activity.

Deploy and Refine Models at Speed

Pricing agility is central to the MGA model. hx Renew’s modular design and low-code environment allow pricing models to be built, tested, and deployed in days rather than months. This flexibility enables MGAs to adjust rating factors, launch products, or shift appetites without waiting on constrained IT resources. Faster model build leads to quicker market response, which capacity providers expect.

Actuaries are often tied down by code optimisation and maintenance. The Actuarial Agent supports them as an intelligent copilot trained on the hx platform. It generates schemas, builds integrations, and explains complex model logic in plain English. Actuarial teams can focus on pricing strategy and innovation instead of repetitive technical tasks.

Rate Adequacy

Rate adequacy defines credibility. A deal may appear profitable on its own, but without portfolio visibility, underpricing can build unnoticed. For many MGAs, portfolio review remains quarterly or annual, assembled in spreadsheets after market conditions have shifted. The chance to act has already passed.

Batch rating and scenario testing are widely discussed but difficult to execute. Disconnected systems make alignment between underwriters, actuaries, and developers complex. Models are difficult to collaborate on. Hours are consumed by cleansing and reformatting data. When re-ratings are produced, they often reduce to columns of figures without policy context.

As a result, portfolio intelligence is often deprioritised in favour of immediate model updates. MGAs are left reacting rather than directing, with limited ability to prove to capacity providers that they are protecting rate adequacy.

What is needed is portfolio insight at the point of pricing. Systems should allow bound and unbound policies to be re-rated, assumptions to be tested, and scenarios to be run without fragile spreadsheets or scarce developer support. This shifts the conversation. Instead of focusing only on individual risks, MGAs can show capacity providers how strategy plays out across the book.

Spot and Adapt to Emerging Risks

For MGAs, how data is captured determines how quickly they can respond to market shifts. When submissions arrive fragmented, time is lost in clean-up and opportunities to identify trends are missed.

Capturing data within the pricing process ensures it is structured, reliable, and immediately available. MGAs can then spot emerging risks, adjust products in line with broker feedback, and demonstrate to capacity providers that they manage portfolios with discipline.

Conclusion

MGAs must deliver results with limited resources while maintaining credibility with brokers and capacity providers. To do this effectively, they need more than isolated tools. A strong foundational technology stack, including a pricing and underwriting platform, is essential. Data must be captured consistently, stored reliably, and kept clean through ingestion processes such as semantic mapping. It then needs to be surfaced directly to underwriters at the point of pricing, ensuring decisions are informed by live context. At the same time, MGAs require the flexibility to build and adapt capabilities quickly. Only when these elements work together—collection, hygiene, accessibility, and adaptability—does AI create real value. Technology must operate in sync, avoiding reliance on point solutions, to deliver lasting impact.

See how these capabilities work in practice:

• Explore Ingestion Agent

• Watch the Actuarial Agent demo