For most underwriting teams, the submission process looks roughly the same as it did a decade ago. A broker sends in a risk. Someone reads it, extracts the relevant data, fills in a template, chases the broker if something is missing, and eventually re-keys everything into the policy administration system. At every step, a person is doing work that doesn’t require underwriting judgment, it just requires time.
Insurance submission automation is the practice of replacing those manual steps with software. Rather than an underwriter reading and transcribing a submission, a system reads it, extracts the data, maps it into the right template, follows up on gaps, and pushes the completed record into the PAS, without manual intervention at each stage.
The distinction that matters is this: automation handles the process. The underwriter still makes the decision.
Why has it become a priority now
Submission volumes have risen sharply across commercial and speciality lines. Brokers expect faster responses. Underwriting teams, particularly at MGAs with lean headcount, are under pressure to process more without adding staff. According to a 2025 Accenture report, underwriters devoted only 26% of their time to core underwriting tasks in 2024, down from 31% in 2021. The remaining time is spent on administrative work.
That imbalance is what submission automation addresses. When data extraction, template mapping, and system entry happen automatically, underwriters regain time for the work that actually requires their expertise.
What the process looks like in practice
A fully automated submission workflow typically covers four steps:
The first is extraction. The system reads the incoming submission, regardless of its format, and extracts all relevant data. No reading, no copying, no interpretation required from the underwriter.
The second is mapping. The extracted data is populated directly into the underwriting template or form. What previously took 20 to 30 minutes per submission happens in seconds.
The third is completion. If information is missing, the system identifies the gaps and follows up with the broker automatically, by email or by phone, to request what is needed. The underwriter is not involved until the submission is complete.
The fourth is export. Once the record is complete, the data is pushed directly into the policy administration system. No duplicate entry, no transcription errors.
How much automation is the right amount?
The answer varies by team, risk type, and appetite. Most modern submission automation platforms offer a choice between a supervised mode, where the underwriter reviews each step before it progresses, and a fully automated mode where the system handles the end-to-end process and the underwriter steps in only when a decision is required.
Neither is inherently superior. High-volume, lower-complexity submissions may suit full automation. More complex risks, or teams building confidence in a new system, may prefer to review each stage initially.
The important point is that the underwriter retains control. Automation removes the administration, it does not remove the judgement.
Nora by Insly
Nora is Insly’s submission automation layer for MGA and insurance underwriting teams. It covers all four steps, including extraction, template mapping, broker chasing, and PAS export, and operates in both visible and invisible modes, depending on the level of oversight the underwriter wants. It sits on top of existing workflows and integrates with your current system, so there is no need to replace what is already working.
For teams processing significant submission volumes with limited headcount, the efficiency gains are immediate. Three underwriters at one Insly customer now manage £18 million in premium, avoiding the need for 12 additional hires.
If your underwriting team is spending the majority of its time on process rather than decisions, submission automation is where to start.
