Turns Snowflake query-history exports into a finance-readable guardrail: which queries are burning credits, which warehouses are over-provisioned, which owners need tagging hygiene, and what should be remediated before the next spend review.
Which queries are turning warehouse power into avoidable spend?
Snowflake Query Credit Guardrail converts exported query-history rows into owner-visible remediation lanes for cache reuse, warehouse sizing, tagging hygiene, and spend-review sequencing.
Queries7
Credits216.7
Avoidable65.26
High risk2
Data-platform spend becomes a GTM problem when reporting teams cannot explain margin leakage, customer-facing analytics slow down, or sales and success teams lose trust in warehouse-backed dashboards. This surface translates technical waste into operating risk.
The value is not just lower Snowflake spend. It is a repeatable cost-governance motion: recover avoidable credits, protect board reporting, assign owners, and convert raw query history into a credible savings narrative.
The repo ships a credential-free CLI, JSON and markdown output, SQL extraction template, deterministic fixture analysis, unit tests, static site generation, and a CI safety scan for local usernames and deploy secrets.
Like the broader Kinetic Gain estate, this repo converts hidden operational drag into named lanes, evidence, owner accountability, and a board-readable next action. Here the drag is Snowflake credit burn and query hygiene.
Query findings
HIGH / 99
62.4 credits
qh_margin_rollup_001
FINANCE_WH / finops
Convert repeated scans into cached, incremental, or materialized-result paths.
HIGH / 99
58.9 credits
qh_margin_rollup_001
FINANCE_WH / finops
Convert repeated scans into cached, incremental, or materialized-result paths.
MEDIUM / 52
41.5 credits
qh_daily_model_102
ML_WH / data-science
Downshift warehouse size or move the query to a smaller governed lane.
MEDIUM / 52
16.7 credits
qh_adhoc_889
ANALYST_WH / unassigned
Downshift warehouse size or move the query to a smaller governed lane.
MEDIUM / 38
24.2 credits
qh_customer_360_217
REVOPS_WH / revops
Review scan pruning, clustering fit, and warehouse scheduling.
Pressure map
Primary recommendation: Start with FINANCE_WH: Convert repeated scans into cached, incremental, or materialized-result paths. This lane has 62.4 credits tied to very high credit burn, large scan volume.
Warehouse pressure
- FINANCE_WH121.3 creditsrisk 99.0
- ML_WH41.5 creditsrisk 52.0
- ANALYST_WH16.7 creditsrisk 52.0
- REVOPS_WH24.2 creditsrisk 38.0
- PIPELINE_WH9.8 creditsrisk 20.0
Owner pressure
- finops121.3 creditsrisk 99.0
- data-science41.5 creditsrisk 52.0
- unassigned16.7 creditsrisk 52.0
- revops24.2 creditsrisk 38.0
- data-platform9.8 creditsrisk 20.0