Smart Data Solutions
Smart Data Solutions runs claims, authorizations, appeals, enrollment and correspondence work for health plans, third party administrators and providers on what it calls an AI native platform, relaunched in July 2026, that joins intake, decision intelligence and workflow orchestration. Digital Mailroom turns paper, fax and electronic submissions into structured records using optical character recognition, language processing and machine learning, and the clearinghouse validates and routes EDI 837 and 835 transactions, eligibility checks and attachments between providers and payers. Document Intelligence organizes records for claims and utilization management staff, Intelligent Connectivity handles portals, APIs and FHIR exchange, and Intelligent Operations adds staff who resolve exceptions and complex work around the clock across the Americas, EMEA and Asia Pacific.
The platform feeds payer systems such as Facets, QNXT and Javelina along with utilization management tools, and a prior authorization offering with Basys.ai pairs Smart Data's intake and record handling with Basys.ai's policy matching. Network Health used the clearinghouse to restore its transaction flow after a 2024 outage, Health Admins, formerly Crown Administrators, cut medical claim review time by 45%, and one customer saved more than $300,000 a year by eliminating paper claims.
Founded in 2000 and based in Eagan, Minnesota, the company reports more than 600 payer, provider and network customers, opened an innovation lab in Phoenix in November 2026 and holds HITRUST r2 certification. Analysts at Everest Group and Avasant placed it in their 2026 payer operations assessments, and pricing is scoped by module and workflow.
Capability Axes
An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read
Machine learning classifies incoming documents and extracts structured data for claims, appeals, enrollment and correspondence, and Document Intelligence builds reviews and summaries on those records. Since the July 2026 relaunch, the company describes the whole platform as AI native, with vision language models in its intake architecture.
Large parts still run on standards and rules. The clearinghouse validates and routes EDI transactions against payer specifications, managed operations staff work exceptions, and the clinical criteria matching in prior authorization comes from partner Basys.ai. The learned functions matter most at intake, where unstructured paper and fax become usable data.
Complete transactions route automatically, and items that need attention are flagged. The claims reconciliation design shows three paths, resolved automatically, flagged, or held in a separate queue, an earlier correspondence workflow added a second review when the model was not confident enough, and managed operations specialists take the complex cases.
Those paths define who handles exceptions and how review starts. Customers are not given an override for every AI assisted action, and nothing establishes that a person approves each output before it moves downstream into a payer system.
Machine learning classifies documents, optical character recognition captures text and natural language processing turns content into fields, while validation rules check transactions and reconciliation results decide how incoming claims are routed. The current architecture refers to large vision language models across three layers, intake, intelligence and orchestration, without naming any release. Chief technology officer Mark Fidow, appointed in December 2025, leads the platform work, and an innovation lab in Phoenix opened in November 2026 to work on problems at the point of intake.
Inputs, outputs and downstream jobs are described for each component. The full model architecture, the training data used and performance thresholds for individual modules are not.
Document capture has used the company's own machine learning since before the current platform, and Basys.ai is the named partner for the clinical policy and authorization engine in the joint prior authorization offering, which Smart Data took on because clinical decisioning is not its core work. That division is clear: Smart Data handles intake, capture and data management, and Basys.ai evaluates coverage rules.
The vision language models in the current platform are not attributed to any provider, and no inventory gives first party and outside model providers, releases, hosting endpoints or training provenance by module.
In a July 2024 customer account, Crown Administrators, since renamed Health Admins, reported a 45% cut in medical claim review time after adopting the document tools, naming the team, workflow and measure but no sample size or comparison group. Network Health restored its clearinghouse connection within days in 2024 after its previous service failed, and a current case study reports more than $300,000 a year saved by eliminating paper claims.
These cases show real operational use of the document and connectivity products. None isolates the effect of the learned models or measures claim decision accuracy under the current platform.
The privacy notice covers healthcare claims and records handled through the portal, APIs and file exchange. An earlier capture workflow, SmartZone Snippets, showed staff only individual areas of a claim for manual validation while documents stayed in controlled facilities, and the current platform emphasizes traceability of every transformation and exception.
Those controls describe how sensitive documents move through parts of the service. Whether collected claims can be reused for training across current models and partners, and how long extracted fields, attachments and review records are kept before deletion, are not set out.
Claims and medical records pass through the clearinghouse and workflow services, the privacy notice addresses that data and names a compliance contact, and the company describes healthcare privacy controls for document processing. Its work for government and commercial payers depends on protected information.
That dependence does not show which entity signs an agreement for each service. No business associate agreement is published, nothing says which plans include one, and nothing states that every model partner and managed operations subcontractor, including staff in its international delivery regions, signs equivalent terms.
HITRUST CSF certification was announced in 2021, and the site now shows HITRUST r2 and an AICPA SOC 2 Type II mark. A document capture account also named SOC 2 and described controlled facilities, and the current platform describes an architecture aligned with those frameworks.
No current HITRUST certificate with its tested scope, and no SOC report showing the auditor and period, can be retrieved publicly. With processing spread across regions and partners, which production systems, facilities and outside components the assessments cover is the open question.
The clearinghouse and document products organize records, check claims and route transactions in payment workflows, with no tool for diagnosing or treating a patient. Prior authorization endpoints, FHIR and attachment support are framed around the CMS interoperability and prior authorization rule and its 2027 deadline, and the Basys.ai offering supports machine assisted review of coverage criteria by health plan staff.
Those references describe the administrative regime the work falls under. They do not amount to FDA clearance, and no independent evaluation confirms compliance with every requirement for a particular payer installation.
The platform keeps records of data transformations, AI assisted decisions and downstream actions, and configuration can differ by payer, line of business and workflow, with transactions set aside for extra review when needed. An earlier capture workflow sent documents the model could not read confidently to separate validation, and a customer council, the S.M.A.R.T. Council, gives clients a say in product direction.
No performance is broken out by member population, payer or document type, and no independent audit has examined bias in approvals or denials that pass through the joint authorization product with Basys.ai.
Exceptions go to operational queues, and the platform records how data entered, changed and moved through each workflow, so a payer can reconstruct what happened to a document or claim. A support line handles service problems, and payer or clinical reviewers look at flagged items.
The headline accuracy score on the site carries no method and no claims or authorization denominator. No contractual correction deadline, refund or indemnity covers mistakes in extracted documents, claim routing or authorization support, so remedies depend on the negotiated service agreement.
The clearinghouse handles electronic claims, eligibility, remittance advice, enrollment and attachments, and intake accepts EDI 837 and 835 files, fax pages and FHIR API calls. Output flows into payer platforms such as Facets, QNXT and Javelina and into utilization management systems, and the platform supports FHIR endpoints and prior authorization APIs for the CMS interoperability rules. Network Health is a named production clearinghouse customer.
The integration role is payer transaction routing. No named EHR appears in a deployment, and nothing shows the company reading from or writing to a clinical chart in production.
The company runs managed data processing, portals and a transaction network. An earlier document capture account said claim data in its SmartZone Snippets workflow stayed in domestic processing facilities, while the current site describes delivery regions across the Americas, EMEA and Asia Pacific supporting round the clock operations.
That global footprint makes the location of each workflow a real question. No data location is given for every hosted service, AI endpoint and backup, no region list is offered, the joint prior authorization service has no residency commitment, and there is no customer owned deployment of the current platform.
Sales go through a demonstration or enterprise inquiry. Buyers choose among document capture, clearinghouse connectivity, document analysis, connectivity services and managed operations, and the modular platform lets them add functions and set how work passes to their own systems and staff.
No subscription price, rate per transaction or minimum volume is published, and the offer does not say which configuration, data migration, custom integration or managed staffing is included. Providers submitting claims through the Quick Claim portal deal with a different side of the business from the payers who buy the platform.
Health plans, third party administrators, providers and health technology partners use the platform for claims intake, enrollment, correspondence and authorization operations. Network Health uses the clearinghouse, Health Admins uses document processing in its claims review team, and the company also supports Medicaid and Medicare Advantage correspondence, provider enrollment and arrangements for technology partners.
Work is configured by contract, rules and input format, and managed operations take on complex records and exceptions for teams that want staff support. The more than 600 customers span payers, providers and networks, though few are named.
Alternatives to Smart Data Solutions
Other products that do the same kind of work as Smart Data Solutions, and what each one does.
Intelligent Utilization Management is a prior authorization product from Availity, the Jacksonville company behind the health information network that health plans and providers use for eligibility, claims and authorizations.
basys.ai develops AI software for prior authorization, utilization management, clinical auditing, and payment integrity.
Cohere Health sells a clinical intelligence platform, Cohere Unify, to health plans. Its modules cover utilization management, payment integrity, appeals, care management, claims operations and HEDIS quality work.
Prior authorization, appeals and grievances platform for health plans and pharmacy benefit managers, across medical and pharmacy benefits.
Silna runs prior authorizations, benefit checks and insurance monitoring for specialty provider groups on what it calls a Care Readiness Platform.
Latent Health sells pharmacy intelligence software that helps health systems get patients onto specialty, infusion and retail medications faster.
Pricing
Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.
| Entry Price | Pricing Basis | BAA Tier | Implementation | Source |
|---|---|---|---|---|
|
Custom quote
|
The commercial offer is scoped around the selected modules, transaction flows and managed operations. No recurring price or rate per transaction is published. | The company handles protected claims and related records and describes HIPAA controls. Public materials do not identify the business associate agreement tier or price. | The company configures integrations, routes and processes for each customer, with setup and integration charges set in the proposal. | Vendor Published |
Customers can scope Digital Mailroom, Clearinghouse, Document Intelligence, Intelligent Connectivity and managed operations in one proposal. Rates and limits for individual components are not published.