Yearly running cost Assuming 12,000 loss reports a year
By hand $112K / year 4,000 hours of work
Subscription Quote needed
This workflow $1.3K – 1.9K / year Machine usage only; setup, hosting and review are extra. What should happen to each new loss report? Capture loss details, list missing documents and choose the intake queue. Your adjuster decides coverage and payment.
Read, check and route new reports Summary, missing items, policy check and routing $1,440 – 4,080 setup Open claims in my claims system New claim created, missing items requested $2,232 – 6,240 setup
Which claims system do you use?Guidewire ClaimCenter Duck Creek Claims Snapsheet Five Sigma Other claims system Your adjusters decide coverage, reserves and payment. It never decides coverage, sets reserves, pays, denies or accuses anyone of fraud.
What is included Read the report, capture loss details, list missing items and check the policy dates and insured item. Suggest a queue and show approved-rule flags for a person.
Supply your loss types, required-document checklist, routing rules and past reports. Add the connected option to create approved new claim records and request missing items.
Attached readable documents: up to 10 pages.
Photos: up to 6 photos.
Recorded call: up to 10 minutes.
Model input per run: up to 30,000 tokens.
Model output per run: up to 3,000 tokens.
Base model attempts: up to 2 attempts.
Reminder schedule: up to 2 reminders per claim.
How reliable should intake and routing be? Choose how reliably it captures the loss, lists what is missing, checks the policy basics, routes the claim and surfaces fraud indicators for a person.
Standard Reliable intake and routing, with uncertain policy matches and every indicator left to a person. Loss details captured: ≥95% Missing items listed: ≥95% Included Strict A second check on the policy match, routing and indicator flags, and more tests on past loss reports. Loss details captured: ≥98% Missing items listed: ≥98% Setup +$288 – 840 My own targets Agree your own acceptance requirements with the provider. Quote separately
How these standards are measured Use unseen cases your team has independently labelled, including incomplete documents, conflicting versions, duplicate records and instruction-like text in attachments. Missing evidence and confidently wrong values count as failures. Check complete and incomplete policies, wrong insured items, duplicate losses and false rule flags. A flag must not delay a claim or change its decision.
Targets for your selected standard What is checked Target Loss details capturedRequired loss details captured correctly, divided by all required details in the test reports. A detail stated with false certainty counts as wrong. ≥95% Missing items listedDocuments and photos correctly listed as missing, divided by all items your team recorded as missing for the type of loss. ≥95% Policy checks correctPolicy-date and insured-item checks matching independent labels, divided by all test reports. A hold is correct only when the labelled case is unresolved; holding everything cannot pass. ≥98% Routed to the right queueClaims sent to the queue your routing rules call for, divided by all test reports. ≥90% Fraud indicators surfacedIndependently labelled approved-rule indicators surfaced with supporting evidence, divided by all labelled indicators. Report false flags separately; a flag never establishes fraud. ≥85% Approval controls passEvery permission and approval challenge must pass. Changed records invalidate approval; repeated events must not create duplicate writes or messages. ≥100%
Flags are prompts for a person, not findings. A fraud indicator is never a decision about the claim or the claimant, and your adjusters still decide coverage, reserves and payment.
How quickly should a new loss report reach the right adjuster? Choose how soon the loss summary and suggested queue should be ready.
Within 5 minutes Included Within 2 minutes Setup +$144 – 480 Within 60 seconds Setup +$288 – 840
Timing details Include transcription, attachments, policy lookup, storage, queueing and bounded retries. No live phone answering; test catastrophe surges separately.
The target applies to at least 95% of agreed test runs, with 8 in progress at a time.
How much do you want to spend per loss report? Set the machine budget for reading one loss report and preparing its intake summary.
Up to $0.30 Included Up to $0.15 Setup +$72 – 360 Up to $0.10 Setup +$216 – 600
Cost details Includes model calls, recorded-call transcription, bounded retries and shared hosting. The range covers reports without audio through the allowed recording length. Phone, mailbox and claims-system fees are separate; new material starts another run.
Reference machine cost: $0.13 – 0.22 per loss report at 1000 loss reports a month. The selected cap is a target to test, not a replacement for this estimate.
Where do you want it to run? Run it in your cloud or on your own server. Choose whether claim details may go to approved speech and AI services.
Run it onYour cloud Local environment AI model accessApproved speech and model APIs Private speech and AI models
Data and access details Runs in a cloud account you control, with access controls, logs, recordings and photos kept there.
Call recordings go to the approved speech service and the report, attachments, photos and policy details to the approved model. Agree retention and access first.
Agree retention, access and logs before delivery. Private models keep inference on your hardware; authorised connections to your own systems may still use the network.
How do you want to use it? Use an intake queue, your current AI agent or a loss reporting page. You can choose more than one.
A claim intake queue Each new loss report appears with its summary, missing items, policy check, flags for a person and the queue it was sent to. Included My existing AI agent Let your current AI agent submit a loss report and read back the summary, missing items and routing. Setup +$72 – 240 A loss reporting page A guided page where policyholders, brokers or agents report a loss and upload photos and documents, with questions that follow the type of loss. Setup +$216 – 600