June 16, 2026
CASE STUDY
MMMM D, YYYY
xx min read

The Convr Risk Context Engine: Why it Matters to the Chief Underwriting Officer

The problem every Chief Underwriting Officer (CUO) faces isn't a shortage of AI tools. It's a shortage of AI they can trust and embed seamlessly within their workflow.

Generative models, agentic assistants, and large language models have proliferated across the insurance industry at an unprecedented pace. Nearly all of them share the same critical flaw: they’re built on general-purpose foundation models that were never trained on commercial insurance and have never seen a real submission, a real loss run, or a real underwriter's decision. They can produce fluent text about underwriting without understanding it. For a CUO responsible for combined ratios, regulatory defensibility, and the consistency of thousands of risk decisions per year, that gap is a material liability.

The Convr Risk Context Engine (RCE) is the answer to that problem and it’s the only answer of its kind.

What makes the RCE unique

Unveiled on June 9, 2026, the RCE is a commercial P&C knowledge graph and semantic ontology that encodes the language, structures, exposures, classifications, and decision logic of underwriting into a unified, machine-readable model, calibrated against a decade of real submissions, real exposures, and real underwriter feedback from leading carriers in production.

The practical implication of that architecture is profound. The RCE does not approximate what a painting contractor is; it knows the difference between a painting contractor and a roofing contractor at the classification level and it knows how that difference should affect appetite, coverage, and pricing. It understands that "general liability for a habitational account in coastal Florida" carries a specific set of exposure signals that have nothing in common with "general liability for a light manufacturing operation in the Midwest." Rather than just pattern recognition over text, the RCE is structured knowledge about commercial insurance, expressed as a machine-readable graph that every AI capability in the Convr workbench runs on top of.

Calibrated on a decade of production data and more than 2,500 integrated sources, the RCE powers every AI capability across the Convr AI Underwriting Workbench from intake to business classification, risk scoring, data enrichment, and workflows.

Why this matters operationally to the CUO

The CUO's mandate is to make good risk decisions, consistently, at scale, in ways the organization can defend and document exactly with reliability. The RCE advances all these dimensions simultaneously.

Consistency: One of the most persistent sources of combined ratio deterioration is inconsistent appetite application . . . underwriters in different territories or teams making materially different decisions on similar risks. The RCE delivers consistent, traceable, verifiable risk data, in-line, which means the same exposure in the same class code is evaluated against the same criteria every time, regardless of which underwriter opens the file or which office processes the submission. The CUO sets the appetite rules; the RCE enforces them uniformly.

Defensibility: Every classification, appetite call, and risk-score output traces back through the ontology to the source submission documents, loss data, and underwriter decisions that informed it. The regulatory direction is reinforcing the value of the RCE. When a regulator, reinsurer, or internal audit function asks why a particular account was accepted or declined, the answer is not a probability score from a black box. It is a documented chain of reasoning tied to real data. In a regulatory environment that is increasingly scrutinizing AI-driven decisioning, audit-ready outputs are not a nice-to-have. They are becoming a condition of doing business.

Scale: The RCE reduces submission-through-quote times by 70% and increases new business win rates. Carriers using the Convr AI Underwriting Workbench have documented an 8% combined ratio improvement on commercial auto lines, 20,000 submissions per month processed fully automatically on non-admitted lines, and 38% more quotes generated per underwriting assistant on financial lines, with quote generation time dropping from two hours to 20 minutes. These are not projections. They are outcomes from carriers already running the RCE in production.

The distinction that separates the RCE from everything else

As Convr Chief Executive Officer John Stammen stated at the RCE launch, "Everyone talks about models. The real question is what they're grounded in. Without a commercial P&C knowledge graph and ontology underneath them, generative and agentic AI are confident guessers. The RCE supplies the missing context . . . what a submission means, what an exposure is, what an underwriter decides . . . and turns outputs into decisions a carrier can defend."

That framing captures the CUO's core concern precisely. A CUO does not need AI that sounds right. They need AI that is right, and that can prove it. Convr unifies fragmented insurance data into a structured data model powered by ontology, schema, semantics, and a knowledge graph within the context engine . . . preserving risk relationships and enabling assistive AI to deliver decision-ready underwriting insights and trace those insights to any historical moment in time. The RCE is the infrastructure that transforms raw submission data, third-party enrichment, and historical loss experience into a single, coherent view of a risk, in real time, at the point of decision.

What this means for the CUO's book of business

For small commercial and BOP books, the RCE eliminates the premium leakage and adverse selection that accumulates when submissions are classified by hand. Business classification errors, misapplied territory codes, and underclassed risks . . . the chronic sources of ratio deterioration on high-volume books are caught at intake before they ever reach a rating engine.

For mid-market and multi-line accounts, the RCE compresses the enrichment cycle that consumes the most underwriter time. Rather than spending two to three days researching an account before rating, an underwriter opens the submission to find the business already classified, the exposure already verified, prior loss signals already surfaced, and appetite already scored against the carrier's own guidelines. The judgment call that makes underwriting valuable happens in minutes rather than days.

For large and complex accounts, the RCE provides the CUO with something that has historically been impossible to achieve at scale: a portfolio-level view of risk concentration, exposure accumulation, and appetite consistency across the entire book, updated continuously as new submissions are processed. The CUO who can see the book in real time, rather than waiting for a quarterly report is the CUO who can act on emerging trends before they become loss events.

The bottom line

The Convr Risk Context Engine is the foundational infrastructure that makes AI in commercial underwriting legitimate. It’s grounded in a decade of real production data, structured around the actual language and logic of commercial P&C insurance, and designed to produce outputs that underwriters, CUOs, and regulators can all defend. For the CUO who is already being asked by their board and their reinsurers how they are using AI, and who cannot afford the answer to be "we're experimenting," the RCE is the answer that closes the gap between AI's promise and underwriting's requirements.

‍

blogs

Keep Reading

More articles on AI, underwriting and the future of commercial P&C.

XX MIN READ

The Hidden Cost of Manual Data Entry in Commercial Lines Underwriting

Ask a commercial lines insurance underwriting leader where their team's time goes, and the honest answer often isn't underwriting at all. It's data entry: keying in values from an SOV, cross-referencing a loss run against a submission, retyping limits and named insureds from a PDF into a rating system.

On paper, this looks like a minor operational cost, the price of doing business with documents that don't come in clean formats.  

The visible cost of manual data entry is time. A commercial property submission with a large SOV can take an underwriter or their support staff hours to process by hand, checking property values, occupancy types, and construction details against what's on the application. Multiply that across a full pipeline of submissions, and the hours add up fast.

While time is the cost that's easiest to see, it is often the least significant one. The hidden costs are the ones that don't show up until later.

‍The following are four hidden costs underwriting leaders need to consider:


Cost one: decision quality

Every hour an underwriter spends transcribing data is an hour not spent evaluating it. When manual entry eats into the day, the analysis that should happen around a submission -- spotting a concerning trend in loss history, questioning whether a stated property value is realistic, comparing an account against appetite -- gets compressed into whatever time is left.

Underwriting quality doesn't erode all at once. It erodes in small increments, submission by submission, as the ratio of time spent on data handling to time spent on judgment tips further out of balance. See why this is a hidden cost that cannot be overlooked?

Cost two: accuracy risk

Manual entry is also where errors creep in. A transposed limit, a missed COPE field, an incorrectly keyed TIV, these mistakes are easy to make and hard to catch, especially under volume pressure.

In commercial lines, where pricing and terms often hinge on the accuracy of property and exposure data, a small entry error can compound into a meaningfully mispriced risk. The cost of that error rarely surfaces immediately. It surfaces later, at claim time or renewal, when it's far more expensive to fix.

Cost three: inconsistent turnaround time

Manual processes don't scale evenly. When submission volume spikes, whether from a hard market, a new distribution partnership, or seasonal patterns, teams reliant on manual data entry hit a ceiling fast. Turnaround times stretch, brokers wait longer for quotes, and the accounts that move fastest aren't necessarily the best risks. In fact they're often the ones with the simplest paperwork. That's not a formula for disciplined underwriting; it's a formula for favoring ease over quality.  

Cost four: talent and turnover

There's also a cost that's harder to quantify but increasingly difficult to ignore: the toll manual entry takes on the people who are doing it.

Underwriters and underwriting assistants who spend a disproportionate share of their day on repetitive transcription rather than analysis tend to disengage from work that should be intellectually demanding. In a competitive labor market for underwriting talent, that's a retention risk hiding in plain sight.

Why this is solvable now

None of this is a new problem. What's changed is the availability of tools built specifically to solve it. Convr’s structured data ingestion, purpose-built for the ACORD forms, SOVs, and loss runs that make up commercial submissions, can take on the transcription work directly, pulling and validating data with a level of consistency manual entry can't match.

That data then feeds the Risk Context Engine, Convr’s ontology for commercial P&C risk, so a submission doesn’t sit as an orphaned record, it’s tied to the broader risk picture connected to prior submissions, relationships, and appetite history rather than evaluated on its own. That doesn't remove underwriters from the process. It removes the bottleneck standing between a submission arriving and a qualified underwriter evaluating it.

Rethinking where the real cost sits

The instinct to treat manual data entry as an operational cost is understandable.

It doesn't show up as a line item the way software or headcount does. But its true cost is distributed across decision quality, accuracy, turnaround times, and talent retention, all of which matter far more to a commercial lines book than the hours spent on submission entry alone.

The teams that recognize this are the ones rethinking where their underwriters' time really belongs, and building workflows that let judgment, not transcription, define how a submission gets handled.

If you’re ready for a conversation about re-envisioning how your team can improve the underwriting experience while avoiding some of these time sucking hidden costs, visit convr.com and book a demo today.

‍

XX MIN READ

Convr Prioritizes Communication in Underwriting Workbench

Email

Convr is making it easier than ever to communicate about submissions within the Convr AI Underwriting Workbench. Now there is an email capability where Convr customers can create new messages for submissions. A user would first need to have a specific submission open within the platform to see the email functionality available to them.

Within the left-hand pane they would just need to click, “Email” then “Create New Message.”

From there the “From” section will automatically be generated with their user email and they would need to plug in a recipient email address. The subject line would also be prepopulated with the submission name.

Convr users can also upload submissions assets and additional attachments about the submission in addition to crafting a customized message about the submission.

Comments
Within the Summary screen you can now also add a “Comment” about a submission, and they can be posted anywhere into Forms, Assets, Emails, etc. to support collaboration. Additionally, you can build a thread of comments. You can also reply to your own comment or react to another user’s comment with a thumbs up, as well.

The idea is that you’re creating a record or recorded conversation allowing another user to enter the platform, to get up to speed on the submission chat and join the conversation with the addition of new comments, which will show up within the feed as well.

You can tag users too, so they receive an in-app notification and email. You can also see in-app alerts, click on them and be taken directly to where you as a user were mentioned within the submission. This global, in-app notification feature is useful if a user wants to bring a team member’s attention to a given item within a submission.


The intent is to open lines of communication between underwriting team members to ensure there is greater transparency and oversight of submissions.

Convr is invested in improving the Convr AI Underwriting Workbench user interface for customers and believes these two new communication features will enhance collaboration and visibility throughout the submission process.  
 

To learn more about Emails and Comments capabilities reach out to Convr at convr.com to book a demo.

‍

XX MIN READ

Agentic AI Doesn’t Just Assist, It Acts

For most of its early history, the Artificial Intelligence (AI) that was used in commercial P&C insurance was a co-pilot. It was an always-on analyst sitting beside the underwriting team surfacing data, flagging anomalies, organizing submissions and more. It was genuinely valuable, yet it still relied on a human to make the call.

Agentic AI changes that equation entirely. It doesn't wait for a prompt or pass-back to a human for every decision. It perceives, reasons, decides, and acts autonomously, within defined parameters, at an unmatched speed and scale. With Agentic AI and the organizational shift from AI experimentation to real‑world execution, new challenges are emerging. If Agentic AI systems are making decisions and taking actions, insurance underwriting teams need to be ready. That means new roles and levels of authority need to be defined. That way, Agentic AI agents will operate within clear boundaries, stay anchored to trusted enterprise data, and scale confidently across the organization, so innovation accelerates without sacrificing governance and control.

The reason Agentic AI requires new operational control is specific to its potential independence of reasoning, decisioning and action. It’s collecting information and getting back to the underwriting team member(s) with a result or response. You're no longer just asking it a question, but giving it the autonomy to perform an action — giving it more authority to operate on your behalf.

When engaged via the Convr AI Underwriting Workbench, your organization benefits from the power of  this reasoning capability within an underwriting workflow. For example, it can act on your behalf sending emails back to a broker for more information. But better still, you benefit from the controls required to customize the workflows to your specific business and governance requirements.

What separates Agentic AI from Assistive AI

Assistive AI is reactive. If you ask it a question, you’ll get an answer. If you feed it a commercial insurance underwriting submission, then you’ll get a summary. It’s powerful precisely because it reduces cognitive load — but the human remains in the loop at every decision point.

Agentic AI is proactive. It doesn't wait to be asked or given a prompt. Given a goal — clear a referral queue, flag a declination, prepare a financial analysis — an agentic system executes the full workflow: gathering relevant data, applying business logic, taking action, and reporting the outcome back to the underwriting team.

Here’s a helpful breakdown:



To learn more about Convr’s Agentic AI capabilities and what we’re doing for customers – get a demo now or read more about it on our newly revamped website at convr.com.


‍

Realize End-to-End Underwriting Excellence with Convr AI

Experience how commercial P&C insurance organizations benefit from submission through quote with a frictionless process enriched by AI decisioning, empowering them to make better decisions, faster.