How to extract data from a PDF submission automatically

By Max Pepe, CMO, Cytora

Welcome to Cytora Editions.

Cytora Editions is an ongoing series of short videos lifting the lid on our product, and demonstrating how you can use Digital Risk Processing to accelerate workflows, unlock underwriting capacity and drive premium growth from the right kind of risks.

Episode 1: Extract key data points from a submission.

On this first episode, I’ll show you how to digitally process a risk submission, zooming in on one specific part of the process – how to extract key data points from a PDF automatically, so an underwriter doesn’t have to be involved in processing the risk, and can purely focus on decision making. 

This is a crucial step for any commercial insurer who embraces the importance of transitioning into a digital first insurer. But, it’s always been very tricky. 

The problem with submissions.

In commercial mid-market insurance, submissions typically come in via email, in a PDF format, as an attachment, that needs to be downloaded and worked on manually.

There are multiple fields within a submission, and no ‘standard’ submission format from brokers. Meaning every submission is very different from one to the next. 

So considerable effort needs to go into analysing and evaluating every PDF, just to understand the potential value of the submission. Whether or not it’s in appetite. How the risk fits in with your portfolio goals. And of course, which underwriter or department is best suited to work on it.

It’s an antiquated process. Skilled underwriters, or an underwriting admin team (depending on how you’re set up), will need to manually extract key data points from the PDF, re-key information into workbenches and systems of record, all before the value add work can begin.

The problem here stems from the fact that when submissions come in, the risk itself isn’t digital. The tools you’re using are digital of course; the submission comes in via email, documents are stored on the cloud in a CRM or PAS. But all the risk information itself is trapped inside the PDF, and is generally inscrutable to machines. The risk isn’t digital.

How can this be solved?

To become a digital first insurer, it’s imperative you’re able to digitally process PDF submissions, and create a digital profile of the risk within.

In this video, I’ll show you how to do exactly that.

If you have any questions, or would like a demo please get in touch here.