Actuarial Workbench: End-to-end actuarial lifecycle

Data Extraction & Ingestion feeds every stage above

Thank you!

We'll be in touch shortly to schedule your demo.

See the Workbench

30-minute walkthrough across the carrier lifecycle, with your filings.

Tesora

The actuarial workbench for the entire carrier product lifecycle.

From filings intelligence to bound policy, on one data spine. Ingest SERFF filings into structured artifacts, fit factors against your book, ship Excel-native raters with audit trails, and feed the same model straight into underwriting and reserving.

The carrier product lifecycle, on one workbench

Five stages, one data spine, one audit trail. The workbench covers the full carrier product lifecycle so the handoffs between actuarial, pricing, and underwriting stay lossless.

Why pricing teams choose Tesora

Spend less time deploying software. Spend more time updating your models.

Frequently asked questions

Common questions from pricing actuaries.

See the workbench across the lifecycle

30-minute walkthrough with one of your filings. We will ingest it, version it, and show you a trauma study.

Tesora

clamp(14px, 1.5vw, 18px)

Competitive Market Research

Turn public filings into structured intelligence in hours.

Pull SERFF filings, public rate manuals, and reference materials into a folder. The agent extracts factor tables, base rates, and rating algorithms into structured, queryable artifacts so your team can review filed material quickly and consistently.

SERFF intake

Drop in a filing PDF or zip, get back a runnable rating model and a test-case set. Every constant traces back to the source page.

Folders of source-of-truth docs

Treaty wording, manual pages, ISO circulars. Indexed, citable, and reusable across analyses.

Ad-hoc actuarial analysis

Ask the assistant to profile filed territory factors, summarize an ILF curve from a state filing, or compare structures across jurisdictions.

Routine playbooks

Schedule a filings scan to re-run on cron or when a new filing lands. Diffs are summarized for your team.

Loss Modeling + Factor Design

Fit factors against your book without a Python team.

Bring your loss triangles and exposure data into the workbench. Use the assistant to fit GLM/credibility-weighted factors, profile experience by segment, and produce a defensible memo of every assumption.

Canonical data sources

Connect policies, claims, exposures, and reference tables once. Re-use across loss modeling and rater builds.

Factor fitting assistant

Severity, frequency, and pure-premium fits with explainable diagnostics. Every factor links to the analysis that produced it.

Provenance everywhere

Each fitted constant is annotated with its source, analysis run, dataset, code revision. Audit-ready by construction.

Memo + chart export

One click to a Word/PDF memo with embedded charts. Hand it to the chief actuary or the reviewer.

Rating Algorithm Deployment

From Excel rater to live API, with a full audit trail.

Shipping a rating model safely takes more work than building one. The workbench versions every model, gates promotions through dev → test → prod, runs trauma studies before deploys, and keeps every historical version live for on-leveling.

Excel-native rating engine

VLOOKUP, INDEX/MATCH, IF, nested calculations. Your actuaries' formulas are the source of truth.

Immutable version history

Every save creates a new version with timestamp, author, and change description. Previous versions are never modified.

Trauma studies before deploy

Run the entire in-force book through proposed rates. See the distribution of changes before anything is filed.

SOX-ready promotions

Dev / test / prod separation, role-based approvals, full event log. A deployment surface a regulator can review.

Underwriting

Submissions and portfolio analytics on the same rating engine.

A submissions queue and a real-time portfolio view that share the rating models you already deployed, so a single quote and a portfolio-wide on-leveling use the exact same code path.

Submissions queue

Track every submission from intake to bind. Quote with the live rater; capture overrides with reason codes.

Portfolio lens

Slice your in-force book by territory, class, effective version. Spot concentration before it bites.

Bind-time governance

Underwriters can't accidentally use a retired version. The rating engine enforces effective dates.

Kalepa / Guidewire ready

Same REST API the back office uses. Front-end vendors plug in once.

Reserving

Close the loop. Reserves and BI on the same data spine.

Once policies, claims, and exposures live in one place, reserving and BI become a thin presentation layer. We start with chain-ladder and Bornhuetter-Ferguson dashboards driven by the same canonical data the rater uses.

Triangles from your data spine

No more Excel exports. Triangles update as claims roll in.

Standard reserving methods

Chain-ladder, Bornhuetter-Ferguson, expected loss ratio. Side-by-side with diagnostics.

BI dashboards

Loss ratio by segment, retention, growth. Pre-built and drillable.

Single source of truth

The same data that priced the policy reserves it and reports it. No reconciliation.

Capital, Reinsurance & Captives

Reinsurance program design, and captive analytics on the same spine.

The same exposures and loss data that drive pricing and reserving also drive capital, reinsurance program design, and captive structuring. Run portfolio-level analytics without a separate data pipeline.

Capital modeling

Aggregate-loss simulations, VaR and TVaR by line, attribution to drivers. Same data spine as pricing.

Reinsurance program design

Score quota-share and excess-of-loss layers against your in-force book. Compare ceded loss ratios under stressed scenarios.

Captives

Funding studies, retention analyses, and program-cost projections for parent-owned and group captives.

Stress + scenario library

Reusable stress scenarios (cat events, inflation shocks, reserve adverse development) applied uniformly across capital, reinsurance, and captive views.

Built for actuaries, not engineers

Excel formulas, factor tables, memos. The artifacts your team already produces, versioned, governed, and live.

Every stage shares one data spine

The folder that fed the competitor scan is the same folder that feeds the rater that feeds the underwriter.

Enterprise-grade security

SOC 2 Type II. 256-bit encryption. Strict role separation across dev / test / prod.

Why organize the product around the carrier lifecycle instead of by feature?

It mirrors how the work flows. Biz dev research feeds loss modeling, which feeds the rater, which feeds underwriting, which feeds reserving. When all five live in one workbench on one data spine, the handoffs stay lossless and the audit trail stays intact end-to-end.

Can we import our existing rating models?

Yes. Upload your Excel raters directly. The system parses formulas, identifies factor tables, and creates a versioned model. Existing logic is preserved exactly.

What integrations do you support?

REST API for any system. Connectors for common policy administration and underwriting platforms. Microsoft 365 (SharePoint and Teams) for document ingestion and notifications.

How does role separation work for SOX?

Three environments: dev (actuaries build), test (QA validates), prod (underwriters use). Different roles have different access. Actuaries cannot push directly to prod. Full audit trail of who promoted what and when.

Do you train on my data?

No. We have zero data retention policies (ZDRPs) in place with our model providers. Your data is not used for model training, ours or anyone else’s.

Is my confidential data protected?

Yes. Each customer is deployed as a single-tenant, bespoke environment that your IT team controls. Your data does not share infrastructure with any other customer.