Submission intake for MGAs and carriers on Guidewire: evaluating your options

Submission intake for MGAs and carriers on Guidewire: evaluating your options

Craig Hangartner

Saba Gobal, CPCU

When a new MGA, broker, or TPA starts sending submissions into your Guidewire environment, someone has to decide how that data gets in. Underwriters then work from whatever arrives, so the intake method sets the quality of everything downstream.

Most teams compare options by demo polish. A better comparison asks where errors enter, who catches them, and what the next sender will cost your team to add.

This article lays out the five realistic intake options for carriers and MGAs on Guidewire. It gives you criteria for judging accuracy, a scorecard you can fill in, and a test plan you can run before you commit.

What are your options for submission intake on Guidewire?

You have five realistic options: manual keying, document extraction, custom connectors, Guidewire’s own add-on tools, and a gateway that accepts structured payloads. Each one moves data from a sender into PolicyCenter. They differ in where the data is structured and who owns the mapping.

Four submission intake methods show how data moves from manual entry and document extraction to custom connectors and gateway integration.

Manual keying from email and spreadsheets

Many programs still start with an email, an ACORD PDF, and a spreadsheet of locations or vehicles. Staff read the attachments and type the values into submission intake screens.

This path needs no integration work. Every field passes through a person’s eyes, but also through a person’s typing, so errors enter at the keyboard.

Document extraction from PDFs and emails

Extraction tools read PDFs and images and convert them to fields. They suit senders who cannot send anything but documents.

The accuracy question here is about the reading step. A scanned form with a handwritten correction or a loss run in an unusual layout produces a field value that someone has to confirm against the page.

Custom connectors per sender

A custom connector takes structured data from one sender and maps it to PolicyCenter. Accuracy is high for that sender because the data never passes through a document.

The cost is repetition. Each sender needs its own connector, and each connector needs maintenance when Guidewire releases change.

Guidewire’s own add-on tools

Guidewire offers intake tooling inside its own platform, built around form-based submissions. It sits close to PolicyCenter and works best when senders deliver the forms it expects.

Do not confuse it with the Guidewire Integration Gateway, which is a developer framework. Yazh (Yaal), the InsOps Integration Gateway, is a separate product that adds a map, test, and send layer beside it.

A gateway that accepts structured payloads

A gateway lets a sender map its payload once, then translates it to the carrier’s Guidewire format. The sender sends structured data, so there is no document to read.

This option moves the work to a one-time mapping step. That makes the quality of the mapping and the review of it the main accuracy questions.

How do you judge accuracy before you buy?

Judge accuracy by tracing where each error enters and who catches it. A platform earns trust through checks you can observe, not through a single accuracy figure. Ask each vendor, and your own team, to show the checks.

A submission accuracy infographic highlights error detection, data validation, incomplete data rejection, and human review of field mappings.

Where errors enter

Errors enter at three points: reading a document, mapping a field, and typing a value. Each option concentrates risk at a different point.

Manual keying carries typing risk. Extraction carries reading risk. Connectors and gateways carry mapping risk, which is one-time and reviewable.

Whether incomplete data is rejected

An intake method that accepts a submission with missing mandatory fields pushes the gap to the underwriter. The better design rejects the payload and tells the sender what is missing.

That is how checking producer codes before data reaches an underwriter works in practice. Clearance problems surface at the door, not in the queue.

Whether a person reviews the mapping

Any tool that uses AI to propose field mappings needs a person to approve them. The review step is what makes the mapping trustworthy.

Put the vendor’s answers next to the questions to ask before trusting AI-assisted field mapping. Weak answers on review and correction are a red flag.

What does each option ask of your team?

Each option asks something different of your engineers, your intake staff, and your underwriters. Count the recurring work, not only the setup. The recurring work decides how many senders you can add.

Engineering effort at setup and after releases

Manual keying and extraction need little Guidewire engineering. Connectors need a build per sender and rework when Guidewire changes.

A gateway moves release handling into the platform. When a payload or a Guidewire release changes, the mapping is updated in one place instead of in every connector.

Onboarding the next sender

Ask what happens when MGA number two arrives. With connectors, the answer is another build. With a gateway, the sender maps its payload, tests it, and sends.

The same pattern covers how MGAs, TPAs, and brokers send submission data into PolicyCenter without building a connector. The sender does the mapping work once, and the carrier does not start a new project.

Where the platform runs and who operates it

Placement affects your operating model. A gateway you operate needs infrastructure and ownership, while a hosted service shifts that work to the vendor.

Weigh where the gateway runs against the skills your platform team already has. Decide this before the pilot, not after.

How do you test options before committing?

Test every option with the same sample set from your real senders. Use messy files, not vendor demo data. Score the results against the same criteria.

Run a sandbox with real sender samples

Collect submissions from three or four senders, including one that regularly arrives incomplete. Send them through each option and record what each one accepts, rejects, and gets wrong.

Yazh gives senders a sandbox and a test harness, so a sender can test a payload before real data moves. Ask every vendor for the equivalent.

Use a side-by-side scorecard

Fill in the table below for each option. Add a row for any criterion your underwriters care about.

Criterion

Manual keying

Document extraction

Custom connector

Guidewire add-on tools

Gateway with structured payloads

Where errors enter

Typing

Reading

Mapping

Reading and mapping

Mapping

Rejects incomplete data

Staff catch it

Flags low confidence

Built per connector

Validation screen

Gap rejection at receipt

Person reviews output

Always

Per document

Developer review

Intake staff

Every mapping

Cost of the next sender

Staff time

Template tuning

New build

Form fit

Sender maps once

Handles Guidewire release changes

Not affected

Partly

Rework per connector

Vendor updates

Updated in one place

Check the full path from account to submission

A submission is more than a form. It needs an account, producer information, and a policy submission that PolicyCenter accepts.

Test whether the option handles an account and its submission in one payload or splits them into separate steps that staff reconcile.

Which option fits your situation?

Fit depends on how many senders you have and what format they send. A carrier with one large partner and a carrier onboarding many MGAs need different answers.

Few senders, documents only

If a small number of senders can only provide documents, extraction or Guidewire’s own tools are the closest match. Plan a confirmation step for every extracted field.

Many senders, or senders that can send data

If you are adding MGAs, brokers, and TPAs regularly, and they can send structured data, a gateway reduces the per-sender build to a mapping and a test. This is the case Yazh is built for.

Carriers on more than one Guidewire product can also look at cross-platform Guidewire solutions. Intake into Guidewire PolicyCenter is the starting point, and the same method extends to other Guidewire applications.

Carry the lesson into claims

Everything above applies to the claims side as well. The same questions about where errors enter, gap rejection, and mapping review apply to claims intake through ClaimCenter.

How InsOps Helps

Yazh (Yaal), the InsOps Integration Gateway, is a data exchange platform for Guidewire. A sender maps its payload once, and Yazh translates it to each carrier’s Guidewire format for PolicyCenter, ClaimCenter, and BillingCenter. LiLa, our insurance-trained model, assists with the mapping, and a person reviews every mapping before it is used.

Senders use a self-service portal, a test harness, and a sandbox to check a payload before real data moves. Incomplete payloads are rejected with the gaps identified, so underwriters work from complete submissions.

Yazh handles changes to a payload or a Guidewire release, supports tenant-specific mapping, and sends large payloads by file location. It connects to Guidewire directly, so submission data flows into PolicyCenter without a custom connector for each sender.

If you are evaluating how to take on the next MGA or broker without a new build, contact us to talk through what this could look like for your operation.

Frequently Asked Questions

What is submission intake on Guidewire?

Submission intake is the process of receiving a new-business submission from an MGA, broker, or agent and creating it as an account and submission in PolicyCenter. It covers data capture, validation, and handoff to underwriting.

Why does the intake method matter for accuracy?

Underwriters decide from the data that arrives, so every error at intake becomes a question later. The method determines where errors enter and who catches them.

How should an MGA or carrier compare intake platforms?

Run the same sample submissions through each option, including messy and incomplete ones. Score each on where errors enter, whether incomplete data is rejected, whether a person reviews the output, and what the next sender costs to add.

What are the common challenges with intake on Guidewire?

Senders use different formats, mandatory fields arrive empty, and connectors need rework when Guidewire releases change. Each new sender repeats the same mapping effort unless you reuse a gateway.

Which metrics show whether intake is working?

Track the share of submissions rejected for missing data at receipt, the number of corrections underwriters make after intake, and the time from receipt to a complete submission in PolicyCenter. Track how many senders you onboard per quarter as a capacity measure.

How does Yazh assist with submission intake?

Yazh assists with mapping a sender’s payload to the carrier’s Guidewire format, and a person reviews every mapping before it is used. It rejects incomplete data and lets senders test in a sandbox first.

How long does it take to bring on a new sender?

The timeline depends on the sender’s payload and the carrier’s mandatory fields. Plan for mapping, a sandbox test with real samples, and a review of rejected payloads before the first live send.

Does a gateway replace Guidewire’s own intake tools?

Not in every case. Guidewire’s tools suit form-based submissions, while a gateway suits senders that can send structured data, and some carriers use both.

Craig Hangartner

Saba Gobal, CPCU