Anavsan Explore
All terms Start Free Trial

Terminology › APEX loop

What is an APEX platform?

APEX is not another cost dashboard. It is the platform-agnostic engine that turns a waste event on Snowflake or BigQuery into owned work, a decision through your review process, and a finding that closes when the metric recovers. Snowflake is the primary worked example.

APEX (Accountability and Performance Enforcement Engine) is Anavsan’s closed-loop system for Snowflake and BigQuery. It builds a Private Knowledge Graph as the ownership substrate — not AutoApply — and runs Trace → Assign → Enforce → Prove so every cost issue has an owner, a proof, and a documented outcome. On Snowflake, that includes 200+ metadata signals.

Key takeaways

APEX is an operating engine, not a chart. Detection is an input. The product is what happens after the alert.
Four stages, one loop. Trace finds waste, not spend. Assign names a party or says needs owner. Enforce drives a decision through your PR process. Prove closes the finding when the metric recovers.
Metadata only. On Snowflake, APEX reads ACCOUNT_USAGE and related signals. On BigQuery, INFORMATION_SCHEMA-class metadata. It does not export business table contents.
Two surfaces, one engine. The Enforcement Desk is the ranked worklist. Ask APEX is grounded chat over the same findings and SQL.

Why APEX exists

Snowflake and third-party monitors already show warehouses, spikes, and Query History. Teams still stall: nobody is clearly accountable, a production change feels unsafe, and FinOps cannot prove that last month’s “optimization” reduced credits. APEX is built for that last mile — the part where savings are either realized or quietly forgotten.

It is the product implementation of workload governance on Snowflake and BigQuery. Snowflake Workload Governance is the Snowflake flavor. The discipline can be described without software. APEX is how Anavsan runs it continuously: signals in, owners named, decisions tracked, closures evidenced from the data.

What an APEX platform is not

Monitoring / FinOps dashboardAPEX platform
JobShow what spentClose the issue with an owner and proof
UnitWarehouse, account, alertWorkload, owner, finding, verified delta
AI roleSummarize a chartExplain a finding; arithmetic stays deterministic
ChangeAdvice in a ticketExact fix through your review process — never a silent warehouse alter

Native Snowflake anomaly detection can flag a spike. That is stage one. APEX treats the spike as an input to cost accountability. The Native App is in-account monitoring. APEX is the full-platform loop.

What APEX runs

Signals

200+ Snowflake metadata inputs

Queries, warehouses, storage, Cortex usage, serverless compute, and ownership clues. No business data leaves the account. Typical connect is a read-only role.

Memory

Private Knowledge Graph

Maps users, workloads, patterns, and fix history so assignment is specific to your estate, not an industry-average model. See Private Knowledge Graph.

Work

Enforcement Desk + Ask APEX

The Desk ranks addressable waste with a mechanism, dollar figure, owner, and exact fix. Ask APEX answers from those findings with SQL, assumptions, and an as-of timestamp.

In practice

FinOps opens a short list of waste, not a warehouse leaderboard. Engineering gets a specific change and a human-gated PR instead of “please look at ANALYTICS_WH.” Platform keeps metadata-only access. Product detail and trial live on the APEX product page.

Trace → Assign → Enforce → Prove

The four-stage loop APEX runs: Trace, Assign, Enforce, Prove.

Read →

Enforcement Desk

The waste-ranked worklist with owners and closable status.

Read →

Private Knowledge Graph

Why assignment is specific to your Snowflake or BigQuery estate.

Read →

Workload Governance

The operating discipline APEX operationalizes.

Read →

Frequently asked questions

APEX is Anavsan’s Accountability and Performance Enforcement Engine for Snowflake and BigQuery. It builds a Private Knowledge Graph as the ownership substrate and runs Trace, Assign, Enforce, and Prove so cost issues reach documented closure. On Snowflake, that includes 200+ signals. The PKG does not AutoApply.
No. It uses metadata-only access — query fingerprints, warehouse usage, storage signals, and related ACCOUNT_USAGE views. Table contents do not leave the account.
Native monitoring shows warehouses, thresholds, and spikes. APEX ranks addressable waste, names an evidence-backed owner, tracks the exact change through your review process, and resolves the finding when the metric recovers.
No. The Native App is Snowflake in-account monitoring. APEX is the full platform loop: Trace, Assign, Enforce, Prove. Teams can use both. BigQuery has no Native App claim.
Yes. The same Trace, Assign, Enforce, Prove loop runs on BigQuery — jobs, slots, reservations, and scan waste. Nested pages: cost governance, storage intelligence, and dataset lineage. We do not claim Cortex, a Native App, or a 200+ BigQuery signal catalog.