Lumnion

Modeling is the product, and it is not decorative. Actuaries fit gradient boosted trees, random forests and decision trees, as well as generalized linear and generalized additive models, on the carrier's own portfolio and compare them side by side. Micro segment detection and a real time advisory module run over the portfolio. Remove the models and a deterministic rate engine, a data preparation module and an external data module still remain underneath.

Oversight is built into the tool rather than asserted around it. Before a price is committed, models are compared on a common dataset and every factor gets statistical results, with factor and interaction advice. Impact analysis and scenario simulation follow, and audit traceable reporting comes afterwards. The actuary drives every step, and the product is designed for inspection. The gap is at the last step.

The rate engine is sold on pushing a commercial price decision into the market instantly, and nothing names an approval gate, threshold or enforcement mechanism between a simulated scenario and a live tariff.

The named machinery is substantial and aimed squarely at letting the institution validate the model itself. It covers side by side model comparison on a common dataset, statistical results for all factors and models, and factor and interaction advice. Variable grouping, automated reporting and audit traceable output complete it. The central claim needs a distinction the market blurs.

A proprietary methodology converts the output of opaque models into base prices and coefficients, and no published methodology, fidelity measure or validation of that conversion exists. Converting a black box into coefficients after the fact is not the same thing as fitting an interpretable model in the first place. The difference is whether the conversion itself has been validated. Some vendors generate interpretable models directly. Lumnion explains opaque ones afterwards, and both are sold under the same word.

The customer list is long and the outcome evidence is absent. Seventeen named customers appear on the company's own site, including three global insurance groups and a national insurance sector information and monitoring body. For a company of roughly eighteen people that is a substantial installed base. No outcome is attached to any of them. There is no case study, no quantified result and no executive quoted. The largest names carry the least evidence, so a buyer can verify who bought it and nothing about what it did.

Nothing is published on whether a carrier's portfolio, claims or quote data informs anything beyond that carrier's own models. The specific gap is the external data module. It is sold as supplying statistically aggregated information from trusted sources and names not one of them. For a pricing input that will enter a filed or published tariff, an unnamed data source is a material silence. It is the mirror image of Lumnion's complete disclosure on the algorithm side.

The privacy posture rests on an independently audited ISO 27701 certification with the certificate published, not on a policy page asserting good intentions. The operating entity is German and therefore inside the European regime by construction. There is no data processing agreement, subprocessor list or residency statement. Nothing describes what Lumnion may do with the policy, claims and quote data a carrier loads into the modeling environment.

Lumnion presents its credentials well. Four current certifications are listed, each with the edition year stated, the auditing body named, and the certificate published for download in one click. They are ISO 27001:2022 and ISO 27701, plus IT service management and quality management. Nothing that is not a certification is mixed in, with no statute or framework padded in and no analyst badge alongside.

The open question is whether the certification body is itself accredited by a recognized national accreditation body, since a certificate is only worth the accreditation behind the certifier. Also absent are any SOC report, a trust portal, penetration testing disclosure and any description of the controls themselves.

Lumnion is unlicensed and unsupervised. One customer is a national insurance sector information and monitoring body, a statutory institution, which makes it a notable reference but confers no standing. No supervisory examination, filing approval or sandbox participation is published.

Lumnion sells price optimization against individual customer behavior and publishes nothing on who it disadvantages. Alongside risk modeling it offers commercial optimization and behavioral pricing, described as taking pricing to a personal level through behavioral optimization and named as a product capability. There is no fairness testing, no protected class position and no governance statement.

No recourse route, no allocation of responsibility, nothing beyond standard terms of use. The policyholder priced by a behaviourally optimised tariff has no visibility of the model, the coefficients derived from it, or the external data enriching their record, and no route to any of the three.

The supply chain is essentially fully described. Every algorithm family the platform can run is named on the company's own site, the models are fitted on the customer's own portfolio, and no third party model service sits in the path. The disclosure costs Lumnion nothing. For a classical machine learning vendor the algorithms are public and the training data belongs to the buyer, so the question is cheap to answer. The vendors that cannot answer it are the ones that bought a language model and will not say whose. One input to the price stays undisclosed, because the external data module's sources are unnamed.

Delivery runs through cloud interfaces with an integrated rule and rating engine, so a price change reaches the market without a core release. That is a strangler pattern common among modern pricing vendors, and Lumnion positions it explicitly as connecting to any core system. The platform is listed on the Amazon marketplace as a subscription service. The named partnership with SAP Fioneer, a core insurance and banking platform vendor, is a real integration signal rather than a logo. No named connector, protocol or reference implementation is described.

Cloud service on Amazon infrastructure with a marketplace listing and interfaces for integration, and a certified service management system covering support, with a published commitment to respond within twenty four hours. Nothing published on region, residency, tenancy or hosting options, which is a live gap for a German entity selling to German and Turkish carriers whose supervisors take different positions on where policy data may sit.

No price, tier or unit is published on the company's own site. There are two partial signals. Free modules are offered as an entry point, and the marketplace listing states a recurring monthly subscription billed through the cloud provider, though no figure appears in the public listing. A buyer can start without talking to sales and still cannot budget.

Lumnion serves non life and health carriers in two core markets, Turkey and Germany, and the local entities of global groups are among its named customers. Seventeen named insurers span motor, property, health and life pension, plus a national sector data body. The buyer type is narrow by design. It is the carrier's actuarial and pricing function, not brokers, managing general agents or reinsurers. The geographic base is two markets rather than a global one, on a headcount of roughly eighteen.