> ## Documentation Index
> Fetch the complete documentation index at: https://docs.creditbenchmark.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Data Submission
> Over 40 contributing banks submit entity-level Probability of Default estimates to Credit Benchmark on a monthly or weekly data cadence.
Over **40 contributing banks** submit PD data to Credit Benchmark on a regular cadence. Submissions include obligor-level **1-year through-the-cycle (TTC) PDs** linked to internal ratings for **senior unsecured** exposure in the wholesale/commercial book. We work with each contributor during onboarding to map their files into the core data elements needed for processing and validation.
## Submission Process
Contributors deliver updates at least once per month:
Each delivery contains a current snapshot of risk assessments and the core context needed for consensus calculation:
* Latest **1-year TTC PD** estimates linked to internal ratings
* Coverage across Corporates, Governments, Funds, and Financials
* Scope limited to **senior unsecured** exposure in the commercial/wholesale book
* Agreed data fields and entity identifiers required for processing
Banks deliver through Credit Benchmark's secure production channels — browser-based **CB Secure** or automated **SFTP**. Where needed, CB can connect to a client's existing secure portal or file transfer process agreed during onboarding.
Submissions undergo automated and manual checks as part of our [Data Validation](/methodology/data-processing/data-validation) process to ensure data quality and consistency.
## Delivery Options
| Delivery Method | Typical Use Case |
| --------------------------------- | ---------------------------------------------------------- |
| **CB Secure** | Manual upload via web browser |
| **SFTP** | Automated or scheduled file transfer |
| **Client-managed portal or SFTP** | CB connects to a client's existing secure delivery process |
## Data Collection Scope
Standardized risk measures and entity identifiers to support consensus calculations.
| Data Category | Details |
| ---------------------- | ------------------------------------------------------------ |
| **Entity Identifiers** | Legal name, country, identifiers (e.g., LEI where available) |
| **Risk Measures** | 1-year TTC PD, internal rating, and as-of date |
| **Model Information** | Model or scorecard identifier used to produce the PD |
| **Firmographics** | Entity type and sector/industry classification |
We do not collect sensitive operational details or underlying credit analysis inputs.
| Data Category | Details |
| ------------------------ | ----------------------------------------------------- |
| **Facility Information** | Terms, collateral schedules, covenants |
| **Exposure Data** | Loan size, limits, utilizations |
| **Underlying Inputs** | Financial statement inputs or raw borrower financials |
## PD Acceptance Criteria
| Requirement | Description |
| -------------- | ------------------------------------------------------------------------------------------ |
| **Type** | Obligor-level 1-year through-the-cycle (TTC) PDs |
| **Exposure** | Senior unsecured exposure in wholesale/commercial book |
| **Governance** | Produced under bank's model governance and linked to internal ratings |
| **Standard** | Aligned to regulated TTC conventions (e.g., Basel capital PD or jurisdictional equivalent) |
| Type | Reason |
| ------------------------ | --------------------------------------------------------- |
| **System-generated PDs** | Without governed model output or analyst approval |
| **Facility-level PDs** | Collateralized or asset-backed PDs |
| **Retail PDs** | Personal consumer PDs and products |
| **Point-in-time PDs** | Scenario/stress-only PDs that don't reflect TTC standards |
## Entity Concordance
The consensus process begins with entity resolution — ensuring submissions from multiple banks are matched to the same underlying legal entity before any calculations are performed. This is supported by a dedicated team of **25+ specialists** combining contributor-supplied identifiers with commercial reference data, public registries, and country-specific databases.
The matching process:
* Prioritises direct identifiers such as LEI and ticker where available
* Handles naming variations such as "Apple Inc" versus "Apple INC US"
* Cross-checks multiple reference sources before assigning a match
* Assigns one `CBId` per entity for downstream consensus processing
**Example: resolving Apple Inc. across four banks**
| Bank | Submitted Name | Identifier Provided |
| ------ | ----------------- | ------------------- |
| Bank A | "Apple Inc" | LEI |
| Bank B | "Apple INC US" | LEI |
| Bank C | "Apple Inc. (US)" | Ticker: AAPL |
| Bank D | "APPLE INC" | None |
When identifiers are present we align deterministically. When absent, we review names and supporting reference data before resolving to a single `CBId`.
Once entities are resolved, CB assigns its own industry classifications using our [Industry Classification](/methodology/data-processing/industry-classification) methodology.
## Security and Compliance
Credit Benchmark is certified to the following standards:
* **ISO 27001:2022** — Information Security Management System
* **SOC 2 Type 2** — Security, Availability, and Confidentiality controls
* **SOC 2 Type 1** — Security, Availability, and Confidentiality controls
* **ISO 27017** — Cloud Security Controls
For up-to-date information on security controls and regulatory certifications, visit our [Security and Compliance page](https://compliance.creditbenchmark.com/).