BigQuery editions (Standard, Enterprise, and Enterprise Plus) are the foundation of capacity-based pricing. This post breaks down what each edition actually controls, where the meaningful differences lie, and how to make the right call for your setup before you start building.

You’ve done the math and reservations make sense for your workloads. Now you need to pick a BigQuery edition. The instinct is usually to anchor on price: Standard is 33% cheaper than Enterprise, so it looks like an easy win. Enterprise Plus is 67% more expensive than Enterprise, so it’s easy to dismiss. But the edition choice isn’t primarily a pricing decision: it determines which features are available to your reservation at all. Pick the wrong one and you’ll hit a hard block mid-setup, because switching editions requires deleting and recreating the reservation from scratch.

Learn all about BigQuery Editions and Reservations. Download our white paper:

What does your BigQuery edition choice actually control?

Editions in BigQuery are not just a pricing tier: they define what features are available in your reservation. Some capabilities are hard-blocked below a certain edition, regardless of how many slots you configure.

A few important things to understand before you choose:

• Editions are a property of reservations, not storage. You pick an edition when you create a reservation. Your datasets and tables are unaffected: the edition only determines how compute is billed and which query features are available for assigned workloads.
• The choice is not reversible on the fly. To switch editions on an existing reservation, you must delete and recreate the reservation with the new edition type. Any assignments will need to be reconnected.
• You can mix editions across reservations. It’s valid to run a Standard reservation for development workloads and an Enterprise reservation for production in the same organization.

With that in mind, here’s what each BigQuery edition actually gives you.

BigQuery Standard Edition: what’s included, and what isn’t

Standard is designed for development and experimentation. It is the cheapest entry point into capacity-based pricing with BigQuery, and it deliberately trades features for cost.

What BigQuery Standard includes:

• $0.04/slot-hour in US regions (check pricing for your region on cloud.google.com)
• Autoscaling with a hard cap of 1,600 slots maximum per reservation
• Project-level assignments only
• Support for QUERY and PIPELINE assignment types
• 99.9% monthly SLA

What BigQuery Standard does not include:

• No baseline slots (autoscaling only: no always-on guaranteed capacity)
• No slot commitments (no 1-year or 3-year discounts)
• No support for Customer-managed encryption keys (CMEK)
• No BI Engine query acceleration
• No BigQuery ML
• No continuous queries
• Cannot create or refresh materialized views (can only query existing ones created elsewhere)
• Project-level only: no folder-level or organization-level assignments
• Maximum of 10 reservations per admin project (vs 200 for Enterprise)
• No CONTINUOUS, ML_EXTERNAL, or BACKGROUND assignment types

The 1,600-slot autoscaling ceiling is the most immediately constraining limit for production workloads. A single complex query at peak load can easily consume several hundred slots. At 1,600 slots across an entire reservation, a handful of concurrent heavy queries will queue rather than execute, and there’s no way to raise this limit within Standard.

The missing commitment plans matter for a different reason: Standard offers no path to discounted capacity. There are no 1-year or 3-year commitment options available to save 20% or 40% – so even if Standard’s per-slot-hour rate is lower, you can’t lock in discounts the way the Enterprise edition allows.

Standard is the right choice for dev/test environments, proofs of concept, and short-lived workloads where you need cheap capacity without operational overhead. It is not a viable foundation for production analytics.

BigQuery Enterprise Edition: the right default for production workloads

Enterprise is the most widely used edition for a straightforward reason: it unlocks everything you need to run a production reservation correctly.

What BigQuery Enterprise adds over Standard:

• $0.06/slot-hour in US regions (check pricing for your region on cloud.google.com)
• 99.99% monthly SLA
• Autoscaling constrained only by regional quota, no hard slot ceiling (quotas can be increased via support request)
• Baseline slots: the minimum guaranteed capacity always allocated to a reservation, billed per second
• Slot commitments: 1-year at 20% off, 3-year at 40% off the pay-as-you-go rate
• Full feature set: CMEK, BI Engine acceleration, BigQuery ML, continuous queries, materialized views (create and refresh), search and vector indexes
• All assignment types: QUERY, CONTINUOUS, PIPELINE, ML_EXTERNAL, BACKGROUND
• Folder-level and organization-level assignments
• Up to 200 reservations per admin project

The baseline and commitment support is what most teams are actually looking for when they move to capacity-based pricing. Baseline slots let you guarantee a minimum execution floor for always-on workloads. Commitments let you lock in discounts once you have a stable usage baseline, cutting your effective slot rate by 20-40% in exchange for 1- or 3-year terms.

None of that is available on Standard. If your reservation strategy involves any of those mechanisms, Enterprise is not optional: it’s the only edition that supports them. For the vast majority of production workloads, Enterprise is the right answer. The question isn’t “should I use Enterprise?”, it’s “is there a specific reason to move up to Enterprise Plus?”

BigQuery Enterprise Plus Edition: two features, a 67% price premium

Enterprise Plus costs $0.10/slot-hour in US regions (check pricing for your region on cloud.google.com). It offers the same commitment discounts as Enterprise, but on a higher base rate.

The SLA is identical to Enterprise (99.99%). The feature set is also identical to Enterprise, with two additions:

1. Managed disaster recovery: automated cross-region replication and failover for BigQuery reservations
2. Assured Workloads: compliance controls for regulated data environments (FedRAMP, HIPAA, IL4, and similar frameworks)

If your organization operates under a compliance framework that explicitly requires Assured Workloads support for BigQuery, or if you have DR SLAs that require managed failover, Enterprise Plus is the correct choice. These are real requirements in regulated industries.

For everyone else: if you’re asking whether you need Enterprise Plus, you likely don’t. The teams that need it know exactly why: it’s in their compliance documentation or their DR runbooks.

Source: BigQuery editions documentation. Pricing shown for US regions. Check cloud.google.com for your region.

Which BigQuery edition should you choose? Standard vs Enterprise vs Enterprise Plus

The decision between BigQuery’s editions is simpler than Google’s comparison tables make it look. Here’s a quick rundown:

Use Standard when:

• You’re running dev, test, or staging environments
• You’re prototyping a new workload before committing to production
• The workload is short-lived and doesn’t need guaranteed capacity or commitments

Use Enterprise when:

• You’re running any production analytics workload
• You plan to use baseline slots or slot commitments
• You need CMEK, BI Engine, BigQuery ML, or continuous queries
• You want org-level or folder-level project assignments

Use Enterprise Plus when:

• Your compliance framework (FedRAMP, HIPAA, IL4, etc.) explicitly requires Assured Workloads support for BigQuery
• You have disaster recovery SLAs that require managed cross-region failover

One thing worth flagging before you commit: the edition you choose constrains your entire reservation configuration. Standard caps your autoscaling ceiling and removes baseline and commitment options. Choose the wrong edition early and you’ll need to tear down and rebuild your reservation setup after your projects have already been assigned to it.

How does Rabbit help once you’re on Enterprise?

Choosing the right edition is a one-time decision. What’s ongoing is making sure the features that Enterprise unlocks (baseline slots, autoscaling limits, and slot commitments) are actually configured to save money rather than generate waste.

Rabbit analyzes your BigQuery usage and billing data to help with exactly this. A few of the things it handles:

• Reservation Planner: recommends optimal baseline and max slot settings based on your historical usage patterns. Getting these two numbers right is where most of the savings opportunity lives on Enterprise, and where most teams leave money on the table. Projected savings are typically 20–50%.
• Max Slot Optimizer: caps your autoscaler to prevent over-provisioning. Uncapped autoscaling on Enterprise can quietly run up costs during burst periods; the optimizer applies a historically-informed ceiling with a safety buffer, reducing reservation costs by up to 40%.
• Commitment identification: surfaces opportunities to move from pay-as-you-go slot pricing to 1-year or 3-year commitments once your usage patterns are stable enough to justify locking in.

None of these features are accessible on Standard: they depend on baseline slots and commitment support that only Enterprise provides. Getting the edition right is the prerequisite; Rabbit handles the optimization from there. To figure out what your potential savings amount to, use our BigQuery Savings Calculator:

Moving forward after selecting an edition

Choosing the right edition is the foundation. The next layer of decisions (sizing your baseline, configuring your autoscaling limits, and setting up slot commitments) determines whether your reservation actually saves money or quietly generates waste alongside the queries it runs.

Our white paper walks through the complete setup process: how to size baseline slots, how autoscaling actually behaves inside a reservation, and how to configure commitments without locking in more than your true steady-state usage.

Download the white paper:
How To Get Started With BigQuery Editions and Reservations

This is Part 2 of a 4-part series supporting the white paper. Part 1: Comparing BigQuery Pricing Models: On-demand vs Capacity-based Reservations. Coming next: Part 3 goes into Baseline Slots and Commitments, the two configuration decisions with the highest financial risk in any reservation setup. And in Part 4: BigQuery Reservations: How Does Autoscaling Really Work?, we go deep into how max slots interact with BigQuery’s 60-second billing window.