Estimate Azure Service Bus costs for Basic, Standard, and Premium tiers with messaging operations and tier comparison.
Last verified: May 2026
Basic is the cheapest option for simple point-to-point queues at $0.05 per million operations. Standard adds topics/subscriptions (pub/sub) and starts at $10/month plus $0.80 per million operations. Premium provides dedicated resources through messaging units at $677.44/month per MU, with operations included at no extra charge. Premium also supports messages up to 100 MB and geo-disaster recovery.
Move from Basic to Standard when you need topics/subscriptions for pub/sub messaging patterns. Move from Standard to Premium when you need predictable performance, messages larger than 256 KB, geo-disaster recovery, or when your per-operation costs exceed the flat Premium MU pricing.
Brokered connections are concurrent connections to Service Bus entities. Standard tier includes 1,000 connections; Premium includes 1,000 per messaging unit. Extra connections cost $0.03 per connection per day. AMQP connections are more efficient than HTTP as they multiplex operations over a single connection.
Service Bus is best for enterprise messaging with guaranteed delivery, transactions, and complex routing. Event Grid excels at reactive event-driven architectures with pay-per-event pricing ($0.60 per million events). Event Hubs is optimized for high-throughput telemetry and streaming data ingestion (millions of events/second). Choose based on your messaging pattern: command/transactional (Service Bus), reactive/event (Event Grid), or streaming/telemetry (Event Hubs).
The Azure Service Bus Cost Estimator calculates monthly costs across Basic, Standard, and Premium tiers based on messaging operations, brokered connections, and resource units. Service Bus pricing is complex with per-million-operation charges, relay hours, and Premium tier per-messaging-unit fees. The tool compares all tiers side-by-side with feature availability.
Basic supports queues only. Standard adds topics/subscriptions, sessions, and duplicate detection at a higher per-operation rate. Premium provides dedicated resources (messaging units), messages up to 100 MB, VNet integration, geo-disaster recovery, and predictable performance with a flat hourly rate.
Each send, receive, peek, or management API call counts as one operation. Receiving a batch of messages counts as one operation regardless of batch size. Operations on deferred, scheduled, or dead-lettered messages also count. The first 13 million operations per month are included in the Standard tier base charge.
Use Premium when you need predictable low-latency performance, messages larger than 256 KB, VNet integration for network isolation, geo-disaster recovery for business continuity, or compliance requirements that mandate dedicated infrastructure. Premium is priced per messaging unit hour.
Your microservice fleet uses Service Bus Standard with topic/subscription pub-sub, currently at 50M operations/month = ~$45/month. The estimator shows Premium at 1 MU = $1,100/month (worse). But your team is requesting VNet integration for security compliance, which only Premium supports. Premium becomes the requirement, not the cost optimization. You provision 1 MU Premium with auto-scaling enabled and the same workload uses ~0.4 MU average = effective cost is ~$440/month for the Premium with auto-scale = ~10x cost increase, but compliance requirement is met.
The estimator computes Service Bus cost across tiers: Basic (per-million ops × rate × month), Standard (base monthly fee + per-million ops above 13M included), Premium (messaging units × per-MU-hour rate, flat regardless of operations). It compares all three with your stated operation volume and surfaces the crossover thresholds where each tier becomes cheapest.
Premium tier's flat per-messaging-unit pricing (~$1,100/month for 1 MU) is dramatically more expensive than Standard for low-volume workloads, but Premium becomes cheaper than Standard once you exceed ~30M operations/month. The crossover is the right time to migrate, not before.
Service Bus operations are billed per API call, not per message. A receive call that returns 100 messages in a batch is ONE operation, not 100. Always use ReceiveBatchAsync over individual receives — same message volume, 1/100th the operation count.
Auto-forwarding rules (where messages from one queue/subscription auto-forward to another) count as TWO operations per message — once at ingestion, once at forward. Heavy use of forwarding can double your bill. Question whether the auto-forward chain is necessary or if direct routing would work.
Was this tool helpful?
Disclaimer: This tool runs entirely in your browser. No data is sent to our servers. Always verify outputs before using them in production. AWS, Azure, and GCP are trademarks of their respective owners.