INFUSE Recap: From Linear to Layered Rethinking Insurance Operating Models for 2026

Insurance operating models are under pressure from every direction. Distribution is becoming more complex. Brokers, MGAs, and Insurtech’s are moving faster. Data is arriving from more places, in more formats, with more expectations. As the market looks ahead to 2026, one question is becoming harder to avoid: can traditional underwriting models still keep up?

This was the focus of Send’s latest INFUSE webinar, where host Tony Tarquini was joined by Emma Davies, Founder & CEO of Waystone Consulting, Bijal Patel, Co-Founder of Aurora, and Kristian Feldborg, CTO at OneAdvent.

Together, the panel explored what it really takes to move from a linear, backward-looking operating model to one that is more connected, responsive, and better suited to modern underwriting. It is not just a technology conversation; it’s an operating model conversation, a data conversation, and ultimately a trust conversation.

The market is moving faster than many carriers are built to handle.

One of the strongest themes from the discussion was that innovation is no longer happening only inside the traditional carrier. Brokers, MGAs, and specialist technology firms are building more connected, responsive models around underwriting. In some cases, they are becoming more confident in risk selection and pricing than the carriers behind them.

Emma Davies described the tension well. Lots of insurers are still operating with legacy systems, incomplete data, and workflows that rely heavily on manual intervention and spreadsheets. This makes it harder to absorb better data from partners, harder to create a joined-up view of portfolio performance, and harder for underwriters to know when growth is genuinely healthy versus when it may be masking a problem.

The challenge, then, is not growth itself. It is if the operating model can support it safely. As Emma put it, many insurers are effectively trying to bolt Formula One parts onto a family car. In theory, it should make them faster. In practice, the vehicle underneath is not built for the pace.

Real-time underwriting needs more than a new front end.

Kristian Feldborg reinforced this point from a technology perspective. Insurer systems were designed for a different era. A more periodic, manual model built around a specific distribution channel and then expanded over time with added layers and workarounds.

What insurers need now is something more modular, API-based, and better able to integrate with a wider ecosystem of partners and data sources. The shift is not just about replacing old technology but improving the insurer’s ability to ingest, structure, and act on data in something much closer to real time.

That matters for better distribution management, but also for everything the market now wants to do with AI, automation, and smarter decisioning. If the underlying data is fragmented, delayed, or inconsistent, the value of those technologies will always be limited.

As the panel made clear, you cannot build a better underwriting engine on top of a weak foundation.

Data matters, but trust matters more.

When the conversation turned to data, the panel moved quickly beyond familiar points like quality, consistency, and validation. Emma’s answer was more direct…the defining characteristic of good data is trust.

If underwriters don’t trust the data, they don’t trust the outcome, and they will find a workaround.

This observation landed because it speaks to a broader reality across the market. Even with all the discussion around data transformation, many underwriting environments are still propped up by spreadsheets and side processes because people do not fully trust the system in front of them. Every workaround is a signal. It usually points to a gap in confidence, clarity, or usability that hasn’t been resolved upstream.

The panel pointed out that data problems are not impossible to solve. They are too often framed as purely technical problems. The hard part is agreeing around what a data point means, how it should be used, and how the market aligns around it. Technology can help but trust only emerges when the people, processes, and definitions behind the data are also aligned. 

Better visibility changes the relationship, not just the reporting.

One of the most interesting moments in the discussion came in the context of delegated authority and portfolio oversight. Tony raised the long-standing reality that lots of delegated arrangements reveal their true performance only over time, often through a familiar cycle of early profitability, later claims development, and then either collapse or validation.

According to the panel, more real-time visibility can change those dynamics. Dashboards aren’t the solution on their own; it’s the near real‑time data behind them which leads to a more continuous, collaborative relationship.

Kristian pointed out that when insurers and delegated partners are working from more current data, they can identify drift earlier and have a more constructive conversation sooner. Instead of discovering problems only at renewal or through lagging bordereaux, both sides can react while there is still time to adjust course.

That shift will help move portfolio oversight from a rearview-mirror exercise to something more collaborative and more operationally useful.

Algorithmic underwriting only works when control stays visible.

Bijal brought that idea into the context of algorithmic underwriting, particularly in complex commercial and specialty lines. The attraction of faster turnaround, less manual effort, and the ability to make underwriting decisions at scale is obvious, though speed alone is not enough. Carriers also need confidence they can see what is happening, understand why decisions are being made, and intervene when needed.

Trust in algorithmic underwriting does not come from the model alone. It comes from a framework of real-time monitoring, joined claims and exposure data, clear auditability, and the ability to refine pricing and appetite quickly as signals emerge.

The example shared in the discussion was telling. When portfolio data shows a worsening trend in a specific segment, the goal is not to wait for a quarterly review. It is to respond that same day, adjusting appetite, rules, or pricing logic with enough precision to improve long-term loss ratio performance earlier.

That kind of responsiveness starts to reshape the carrier’s role. It is no longer just handing over the underwriting pen or keeping it locked down. It is about creating an operating model where innovation can sit alongside control, and where visibility is continuous rather than retrospective.

Technology on its own is not the differentiator.

Even with the focus on platforms, algorithms, and modular architecture, the webinar returned to one crucial point…winners will not be defined by technology alone.

Emma’s point on this was especially sharp. The firms moving the fastest aren’t just using better tools. They are also bringing a stronger entrepreneurial mindset, a clearer connection to the customer, and a willingness to rethink how the business works. In her words, it is not enough to put someone behind the wheel of the Formula One car, you also need the right driver.

This is a useful reminder for insurers under pressure to modernize quickly. New technology can create leverage, but it will not automatically create customer closeness, better judgment or decisions. Those still depend on how the business is designed and how leadership chooses to respond to change.

The future belongs to insurers that can adapt in real time.

As the webinar closed, the panelists’ takeaways converged around a common theme.

Emma returned to trust, arguing that lasting change starts when leaders stop asking how to force compliance and start asking why people are working around the process in the first place.

Kristian’s message was equally practical: stop managing relationships in the rearview mirror. The market needs to move toward more collaborative, real-time ways of working across carriers, MGAs, brokers, and technology partners.

Bijal connected both points neatly. In algorithmic underwriting, trust comes from auditability, real-time monitoring, and rapid iteration, not from the model alone.

Together, those perspectives point to a broader reality for the market. The move from linear to real-time is not just a systems upgrade. It is a rethink of how underwriting should operate when data is richer, distribution is more dynamic, and decisions need to be made faster without losing control.

Insurers who get this right won’t just be more efficient, they will be better positioned to adapt safely, build stronger partner relationships, and create underwriting models fit for what comes next.

Watch the full webinar here.