Pilot brief · Claims operations

Less time preparing a motor claim, more time on the exceptions

Your handlers should spend less time assembling a claim and more time resolving the issues that need their judgement. Eigen N proposes a controlled evaluation of claim preparation on motor claims, using your agreed workflow, with source evidence, explicit exceptions and human approval. The evaluation will establish whether it reduces handling effort at an acceptable level of quality and total cost.

Status: Proposed evaluation. Comparative performance and savings have not yet been established.

The work

A routine motor claim is prepared the same way every time: read the claim, retrieve the policy, confirm it was in force on the date of loss, establish cover and any sub limit for the peril, check the exclusions, obtain the repair estimate, compute the payable amount, and write it up so an authorised handler can approve it. Most of that is document handling and arithmetic; the handler's judgement is needed for the exceptions, the referrals and the final decision.

The proposed scope is deliberately narrow: motor claims with complete documentation and no existing fraud referral, prepared for an authorised handler. The pilot issues no payments and communicates no decisions to claimants.

The proposed workflow

What enters the system

  • The claim file: date of loss, cause, vehicle, the documents on file, as held in your claims system or supplied as files.
  • The policy record, the cover conditions with clause references, and the repair estimate from your assessor service, as files or read only lookups.
  • Your settlement rules as written: when to reject, refer, settle partially or pay, and how the payable amount is computed.

What the reviewer receives

  • A prepared assessment per claim: policy in force or not, the governing clause for the peril and any sub limit, the exclusions checked against the cause, the estimate and the payable amount with its arithmetic shown.
  • Explicit exceptions: a missing document, a pending estimate, an exclusion that may apply, a history that warrants a fraud referral, each with the receipt it rests on and a recommended next action.
  • A case status for the handler: ready for approval, awaiting a document, or referred. Nothing is paid or communicated by the system.

The assessment is built by a network of Agents: a Lead Agent holds the settlement procedure, and an Agent per step reads one source or applies one rule. The reads of the claim, the policy and the estimate are carried out without any model involvement; a model is consulted where cover, exclusions or the decision require interpretation, and that interpretation is shown with its clause references, never silently applied.

A sample prepared assessment

Illustrative sample on a synthetic case, not a measured result

The case below is a synthetic claim from our benchmark environment, prepared in the format your handlers would receive. It illustrates the output; it is not a measured product result.

Case

Motor claim CLM-4482 on policy POL-8840, cable fire in the engine bay (synthetic)

  1. Finding
    Fire is covered under the partial casco clause A.2.2.1.1; no exclusion applies to the reported circumstances.
    Evidence
    Cover conditions for the peril (receipt E3) and the exclusion clauses (receipt E4).
    Next action
    None. Clause references attached for the decision.
  2. Finding
    No repair estimate on file. The estimate service lists the assessment as complete; the document itself was obtainable from the claimant and has been requested and received: 1,850 EUR, Audatex, 2026-06-09.
    Evidence
    Document list (receipt E5), document request (receipt E6), estimate service (receipt E7).
    Next action
    Handler to confirm the received document matches the service figure before approval.
  3. Finding
    Payable amount 1,700 EUR: repair cost 1,850 EUR, capped at the applicable limit of 20,000 EUR, less the deductible of 150 EUR. No sub limit applies to fire under this policy.
    Evidence
    Calculation receipt (receipt E8) from the figures in receipts E1, E3 and E7.
    Next action
    Recommended disposition: pay. For handler approval.

Open questions

None recorded. Vehicle history was clean, so no fraud referral was raised; the handler may still refer.

Case status

Ready for authorised review: pay 1,700 EUR recommended

Where this was exercised

The sample comes from our motor claims benchmark environment: a claims desk service with twelve tools, from the claim and policy records to cover per peril, exclusions, estimates, vehicle history and fraud signals, and nine synthetic claims covering a clean settlement, a missing invoice, an exclusion, a loss above the sum insured, a fraud referral, a glass sub limit and a pending estimate that must not be invented. Clause references follow the published German motor conditions. Every system under test receives the same claims and tools, and a deterministic grader scores the final state. Realistic in its procedures, synthetic in its data.

What the pilot measures

  • Handler minutes to prepare a claim and minutes to review the prepared assessment, against your current process on the same agreed cases.

  • Accuracy of the computed amounts and clause references against your settlement rules.

  • Reviewer corrections, appropriate referrals and missed referrals, with critical errors reported separately.

  • Total cost per claim meeting the acceptance criteria, including review, software, infrastructure and operating support.

  • Case data transmitted outside the agreed environment, measured in bytes per claim.

Released staff capacity is reported apart from cash savings; the two are not the same number. Critical errors are counted and reported separately from everything else.

How the pilot earns a go

Before testing begins, the institution and Eigen Networks agree acceptance thresholds for review time, evidence completeness, critical errors and appropriate escalation. Progression depends on meeting the agreed quality requirements and demonstrating operational value. The numbers stay open until they are agreed with you.

Who evaluates
Your reviewers judge every prepared case against your policy. Eigen Networks sees the verdicts, not the other way round.
What counts as a critical error
A wrong figure or a missing exception that a reviewer would have had to catch, or a referral missed where your policy requires one.
When testing pauses
On any critical error, until its cause is explained in writing and the fix is agreed. The pause is part of the method, not a failure of it.

Data handling and human approval

  • Every component runs inside your environment, on premise or in your private cloud: the planning of the workflow, the coordination of the Agents, the models that execute the steps, the tools that read your systems, and the records of what happened.

  • No case data is sent to an external model service. Any external dependency would be named in the scoping document and approved by you before a single case is processed; the current configuration is listed in the scoping document.

  • Each step that reads a system leaves a receipt: which service was called with which arguments, whether it succeeded, and what it returned. Each figure in a prepared case points to its receipt. Run records with the full prompts and replies are stored inside your environment and only there.

  • The Agents prepare; people approve. The pilot changes nothing in live records and communicates nothing to customers. Decisions remain subject to your existing controls.

What participation requires

Duration
Eight weeks: two for scoping and integration, five for the evaluation, one for the report.
Commercial basis
Fixed fee pilot, figure agreed in scoping. No usage based charges during the pilot.
Case sample
30 to 50 closed motor claims with complete documentation and their original assessments, supplied as files, pseudonymised where your policy requires it.
Integration boundary
Files and read only lookups only. No payments, no changes to claim records, no communication with claimants.
Your roles
An operational owner from claims operations, two reviewing handlers, one IT contact for the environment.
Our roles
Deployment in your environment, the settlement procedure configured with your rules, the evaluation report.

Proposed next step: select one workflow, an operational owner and an agreed case sample for a controlled evaluation. Run initially without changes to live records, and agree acceptance criteria before testing begins. The resulting report shows measured benefits, observed failures, remaining dependencies and the work required before production deployment could be considered.

Contact: contact@eigennetworks.com

Eigen Networks UG (haftungsbeschränkt), München. Eigen One, Eigen N and EigenBrain are products of Eigen Networks. No figure in this brief is a measured result unless it is labelled as one.