Data lake
TrustVendor + Snowflake
For data teams who want to analyze vendor risk alongside business data, TrustVendor exports structured vendor risk events and scores to a Snowflake database on a configurable schedule.
How data flows
TrustVendor writes vendor risk records — scores, signals, certification events — to a Snowflake schema via the Snowflake Connector. Records are append-only and retain full history.
How to set up
- 1 Create a TrustVendor service user in your Snowflake account.
- 2 Grant write access to a target schema.
- 3 Configure the Snowflake export under Settings > Data Export in TrustVendor.
- 4 Set export frequency (hourly or daily) and select which entity types to export.
Common questions
What does the Snowflake schema look like?
Tables include vendors, scores (with bitemporal timestamps), certifications, signals, and subprocessors. A reference dbt project is available for standard transformations.
How is the Snowflake export partitioned and what are the cost implications?
Records are written as append-only rows partitioned by the export timestamp, so Snowflake's micro-partitioning handles time-range queries efficiently without manual clustering. The scores table is the smallest — one row per vendor per scoring run — while the signals and artifacts tables grow proportionally to monitoring volume. For most programs under 500 monitored vendors, the daily export generates well under 1 GB of new data, which falls comfortably within Snowflake's compressed storage cost tier.
Does TrustVendor include evidence artifact content in the Snowflake export?
No. The Snowflake export contains structured metadata — scores, claim predicates, signal summaries, certification fields, subprocessor names — but not the raw text content of crawled documents. Evidence payloads can contain PII from vendor-published pages; keeping content out of the export prevents accidental PII replication into your data warehouse and keeps export file sizes predictable.
Can I backfill historical TrustVendor data into Snowflake after first connecting?
Yes. TrustVendor retains claim history indefinitely and score history for 13 months of daily rows before rolling up to weekly. When you first connect the Snowflake export, you can request a historical backfill from TrustVendor support that writes all available history for your monitored vendors into the target schema, giving your dbt models and dashboards a full retrospective view from day one.
Is there a reference dbt project I can use with the Snowflake export?
Yes. TrustVendor publishes a reference dbt project at github.com/trustvendor/dbt-trustvendor that includes staging models, a mart for vendor risk scoring trends, and a certification-expiry model suitable for alerting via dbt Cloud. The project is designed to work with the default table prefix and schema names from the Snowflake export, with documented variables for teams that use a different naming convention.