GCP Pub/Sub Lite vs Pub/Sub cost: when the cheaper tier pays off

Quick answer

Pub/Sub bills per gigabyte of message throughput, around 40 dollars per TiB of data published and delivered, with no capacity to provision, so it scales seamlessly and you pay for exactly what flows. Pub/Sub Lite instead requires you to provision throughput and storage capacity in advance, at a much lower unit cost, so for high, steady, predictable volume it can be several times cheaper. The tradeoff is that Lite requires capacity planning, offers fewer features, and charges for provisioned capacity whether or not you use it. The rule is use standard Pub/Sub for variable or moderate volume and its rich features, and Pub/Sub Lite only for high, steady, predictable throughput where the lower unit cost outweighs the capacity management.

Google offers two managed messaging services with very different cost models. Standard Pub/Sub is fully managed and charges for the data that flows through it, with no capacity to plan. Pub/Sub Lite is a lower-cost tier that asks you to provision capacity yourself in exchange for a much lower unit rate. Choosing between them is a classic managed-convenience-versus-unit-cost decision that hinges on your volume and how predictable it is.

The two cost models

Aspect	Pub/Sub	Pub/Sub Lite
Billing basis	Per GiB of throughput	Provisioned throughput and storage capacity
Approximate rate	About 40 dollars per TiB	Much lower per unit
Capacity planning	None	Required
Idle cost	None	Pay for provisioned capacity

Standard Pub/Sub charges roughly 40 dollars per TiB of message data published and delivered, and it scales automatically with no provisioning. Pub/Sub Lite charges for throughput capacity units and storage you provision in a zone or region, at a substantially lower unit cost, but you must size that capacity and you pay for it whether or not messages flow. So Lite trades the seamless scaling and hands-off operation of standard Pub/Sub for a much lower unit price that you have to manage.

When Lite is cheaper

Pub/Sub Lite can be several times cheaper per unit of throughput than standard Pub/Sub, so for high, steady, predictable message volume it wins clearly on cost. A pipeline moving a large, constant stream of events, where you can accurately provision the needed capacity and keep it well utilized, captures Lite's low unit rate without wasting provisioned capacity. The larger and steadier the volume, the more the unit-cost saving outweighs the effort of capacity planning.

When standard Pub/Sub wins

For variable, spiky, or moderate volume, standard Pub/Sub usually wins despite its higher unit rate, because you pay only for what flows and never for idle provisioned capacity, and you avoid the risk of under- or over-provisioning. Standard Pub/Sub also offers richer features, global topics, more integrations, and simpler operations, that Lite trades away. For most applications whose messaging volume fluctuates, the seamless scaling and features are worth the higher per-unit cost, the same logic as other message queue comparisons.

Choosing between them

Estimate your message throughput and how predictable it is. For high, steady, well-understood volume where you can provision capacity accurately, Pub/Sub Lite's lower unit cost can save substantially. For variable or moderate volume, or where you value the richer features and zero capacity management, standard Pub/Sub is the better choice despite its higher rate. Do not adopt Lite for spiky workloads, where provisioned capacity sits idle or throttles bursts. Price both against the resource catalog so the cheaper service is chosen before the pipeline is built.

FAQ

How do Pub/Sub and Pub/Sub Lite differ on cost?

Standard Pub/Sub bills per gigabyte of message throughput, around 40 dollars per TiB published and delivered, with no capacity to provision. Pub/Sub Lite bills for throughput and storage capacity you provision in advance, at a much lower unit cost, but you pay for that capacity whether or not messages flow. Lite trades seamless scaling for a lower unit price you must manage.

When is Pub/Sub Lite cheaper than standard Pub/Sub?

For high, steady, predictable message volume. Lite can be several times cheaper per unit of throughput, so a pipeline moving a large, constant event stream, where you can provision capacity accurately and keep it well utilized, captures the low unit rate. The larger and steadier the volume, the more the saving outweighs the effort of capacity planning.

When should I use standard Pub/Sub instead of Lite?

For variable, spiky, or moderate volume, and when you value richer features and zero capacity management. Standard Pub/Sub charges only for what flows and never for idle provisioned capacity, avoids the risk of under- or over-provisioning, and offers global topics and more integrations. For most fluctuating workloads, the seamless scaling is worth the higher per-unit cost.

What are the downsides of Pub/Sub Lite?

Lite requires capacity planning, so you must size provisioned throughput and storage and pay for it whether or not you use it, risking waste from over-provisioning or throttling from under-provisioning during bursts. It also offers fewer features than standard Pub/Sub, such as no global topics. These tradeoffs are why Lite suits only high, steady, predictable throughput.

How does C3X help choose between Pub/Sub tiers?

C3X prices your messaging infrastructure from Terraform before you deploy, so the cost of standard Pub/Sub versus provisioned Pub/Sub Lite capacity is visible at design time. That lets you compare the two for your expected throughput and volume predictability, choosing the cheaper service before building the pipeline rather than discovering that provisioned Lite capacity sat idle or that Pub/Sub throughput charges were higher than expected.

What to do next

Compare Pub/Sub and Pub/Sub Lite on cost before you build. C3X reads your Terraform and prices your resources against a live catalog. Start with the quickstart.