Where the data comes from, and how to trust it.

3.1 — Gathering FOCUS datasets

You can gather FOCUS datasets directly from your data generator. Most major providers now publish native FOCUS exports.

Open-source tools the FinOps Foundation supports

• FOCUS Validator — CLI tool that evaluates how conformant your dataset is against the spec’s normative requirements. Currently supports v1.2; v1.3 / v1.4 planned.
• Requirements Analyzer — extracts and structures normative rules so you can explore them and build conformance logic of your own.

Validator limitations to remember

• Some FOCUS fields are conditionally required; the validator can’t always determine if conditions apply.
• It may flag missing fields that aren’t actually required in your situation.
• It may fail to flag fields that should be present for your use case.
• Validating a sample doesn’t imply conformance for the whole dataset.

What if your provider doesn’t support FOCUS?

Talk to your account representative. Express your billing-data needs and expectations. The transition to standardised cost, usage, and billing data requires advocacy across your providers.

3.2 — Overcoming common challenges

Specification vs. dataset

The specification describes how the data should be created — it’s also a reference for understanding the data. The dataset is the real-world implementation in FOCUS-aligned format. Datasets are ideally acquired directly from the data generator, in whatever delivery format they offer (object storage, query-engine table, REST API).

Common challenges and solutions

3.3 — Five real-world migration patterns

Pattern 1 — Parallel run, then cut over (STMicroelectronics)

Treat FOCUS migration like a financial systems migration. Run both pipelines, reconcile until numbers match, then cut over. Separate ingestion from enrichment and allocation so reruns don’t require re-extracting all provider data. Use external allocation keys when reorganisations are frequent — tag-based allocation is fragile in re-orgs. Cultural note: “FinOps is not cost killing.” Balance business need, service level, and optimal cost.

Pattern 2 — Custom pipeline with metric-based allocation (GitLab)

Provider data → FOCUS converter → Snowflake → DBT transformations → Tableau. Allocation built from authoritative systems: Prometheus, Thanos, GitLab product data, internal warehouse. Customer-type dimension (free, paid, internal) added during enrichment. Unit economics published at general availability, not retroactively. Principle: shared platform costs need usage-based allocation, not just account- or label-based.

Pattern 3 — Treat adoption as a formal initiative (Zoom)

Dedicated FOCUS initiative tracked in Jira with quarterly OKRs. Audit existing FinOps practices first — find manual reporting, weak allocation, poor granularity, workflow gaps. Concrete targets: ~95% allocation, 75% utilisation, 85% chargeback accuracy, >90% anomaly visibility.

Pattern 4 — Hybrid comparison (UnitedHealth Group)

Use FOCUS to compare data-centre and cloud spend on an apples-to-apples basis. Network cost models differ radically: data-centre networking is pipe/capacity-based and shared; cloud networking is usage/data-transfer-based with major egress, replication, and managed-service charges. Migration estimates are usually wrong — treat them as directional, monitor actuals from day one, watch for the “double bubble” cost during overlap.

Pattern 5 — Public-sector / brokered cloud (European Parliament)

The Parliament receives downstream chargeback through the European Commission as cloud broker; needed more granular data than the Commission’s dashboard provided. Built FOCUS-based interchange (Athena → FOCUS → S3 → cost tool via REST) instead of a bespoke connector. Lesson: a custom connector solves today’s problem but creates tomorrow’s maintenance burden. A shared standard is reusable, auditable, and economical.

Cross-cutting themes

• Standards protect against bespoke integration debt.
• Tool constraints (file size, file count, supported formats) shape ingestion design.
• Brokered cloud needs broker data — native cloud connectors alone don’t show total cost.
• Coexistence is the norm during transition. Plan for it.
• Alignment grows in importance with complexity — governance and coordination matter more as more contributors and stakeholders join.