Standard - FOCUS

Cloud billing data in one standard format, in the same model as your labor and services.

FOCUS, the FinOps Open Cost and Usage Specification, gives billing data from different providers one shared format, so you can ask the same question of all of them without building a pipeline for each vendor. MagicOrange brings every cloud source in as FOCUS 1.3.
What it is

What FOCUS is

FOCUS is an open specification that normalizes billing datasets across AI, cloud, SaaS, data center and other technology vendors. Twenty-one vendors now generate FOCUS data, including AWS, Microsoft Azure, Google Cloud and Oracle, with Databricks and Snowflake among them.

Before FOCUS, every provider had its own field names and its own idea of an effective cost, a commitment or a region. A multi-cloud team built a pipeline for each one and reconciled them by hand.

Three levels of billing data

The monthly summary invoice is enough to reconcile, and no more. Provider dashboards are flexible and less granular. The detailed report or billing API, updated daily, is where FinOps work happens, and it is the level FOCUS standardizes.

At the sourcefocus.finops.org

How a cloud bill is built

Every cloud charge is usage times a rate

Cost = usage × rate

Every cloud charge resolves to a quantity consumed and a price per unit. Everything else in a billing file is detail about one of those two terms: the resource, the region, the account, the tag, the commitment that changed the rate.

What FOCUS standardizes

How that detail is named, so usage and rate stay comparable across providers.

Standing

Where MagicOrange stands on FOCUS

MagicOrange is a Premier Sponsor and Governing Board member of the FinOps Foundation, the body that stewards FOCUS.

Where it runs

Where FOCUS runs in the platform

FOCUS elementWhere it runs
Native billing ingestionData Studio, included with every capability: 233 connectors, including direct billing API connections to AWS, Azure and Google Cloud, with no data warehouse in between.
FOCUS normalizationEvery hyperscaler bill enters in one standard format, so the same allocation runs across all of them.
Tag mappingAllocate with your tags, without tags or on rules MagicOrange derives, and mix all three in one model.
Sign offA formal gate: data is approved by a named person, with a timestamp, before it enters the model.
Daily allocationFinOps: normalized cloud cost allocated daily, to services, products, business units and customers, instead of monthly after the invoice.
The calculation engineDatabricks-native, so the arithmetic is done by rules and never by a language model.
Token costAI Economics: token usage, model API, agent workload and GPU are shipped standard objects. Cloud-borne AI cost enters through FOCUS today.
Questions this page answers

Questions you can answer from one model

  • Can one allocation run across AWS, Azure and Google Cloud without three pipelines and a reconciliation step?
  • What is the effective cost of a workload after commitments, in the same terms for every provider?
  • Which spend arrived untagged this month, where did the rules send it, and how much is still unallocated?
  • When a provider adds or renames a billing column, what breaks, and who finds out?
  • Can finance reconcile the FOCUS view back to the invoice the provider actually sent?
  • How do token costs from a model API land beside the cloud cost of the infrastructure that served them?
Read next

Related standards

FinOps Framework

The principle that points to FOCUS.

FinOps for AI

The category the token fields will serve.

MOTEF

Where FOCUS is the entry format for the public cloud and AI spend categories.

Bring your billing data in one format and work from one model.

See a multi-cloud bill land in one model.