Last updated: June 2026
To cut Azure Application Gateway and Load Balancer costs, right-size the minimum capacity units and cap autoscale, consolidate multiple applications behind one gateway with host and path routing, choose the cheapest SKU that meets the requirement, and delete idle gateways and orphaned public IPs. The largest single saving is usually collapsing a fleet of per-application gateways, each carrying its own fixed hourly fee, into a few shared ones.
- Application Gateway v2 bills a fixed gateway-hour fee plus capacity units; Load Balancer bills by rules and data processed.
- Use Standard_v2 unless you need the managed web application firewall, which requires WAF_v2.
- One gateway with host and path routing replaces many single-app gateways.
- v1 Application Gateway and Basic Load Balancer are retired; migrate to v2 and Standard.
Azure Application Gateway is a layer 7 load balancer that does application-aware routing, TLS termination, and optional web application firewall, while Azure Load Balancer is a layer 4 service that distributes traffic without inspecting it. Cutting their cost means sizing the gateway to real demand, consolidating where you can, and choosing the cheapest service and SKU that meets the routing requirement, rather than running a premium gateway per application. This article is part of our Azure cluster; the pillar it links up to is the complete guide to Azure cost optimization. Right-sizing and consolidating are Cut steps in our See, Cut, Lock, Run method.
How is Azure Application Gateway priced?
Application Gateway v2 is billed in two parts: a fixed hourly fee for the time the gateway is provisioned and available, plus a capacity unit charge that scales with the compute, persistent connections, and throughput it consumes. As of June 2026, per the Azure Application Gateway pricing documentation, the WAF_v2 SKU adds the managed web application firewall and bills capacity units at a higher rate than Standard_v2. Note that the v1 SKU was retired on April 28, 2026, so only the v2 SKUs remain. Because the fixed fee is charged per gateway regardless of traffic, and capacity units scale with load, the two main levers are how many gateways you run and how low you can set the minimum capacity. Verify the current figures on the linked Azure page before acting, because gateway pricing changes.
| Service and SKU | Billing model | Layer | Best for |
|---|---|---|---|
| Load Balancer Standard | Rules + data processed | Layer 4 | Plain traffic distribution, no app routing |
| Application Gateway Standard_v2 | Gateway-hour + capacity units | Layer 7 | Path routing, TLS offload, no WAF |
| Application Gateway WAF_v2 | Gateway-hour + higher capacity units | Layer 7 | Internet-facing apps needing managed WAF |
| Basic Load Balancer / Gateway v1 | Retired | n/a | Migrate to Standard / v2 |
What is the difference between Application Gateway and Load Balancer cost?
Azure Load Balancer is cheaper than Application Gateway because it operates at layer 4 and does no application-level work. Load Balancer Standard bills by the number of load balancing rules and the volume of data processed, with no fixed gateway-hour fee for application routing, while Application Gateway bills a fixed gateway-hour fee plus capacity units for the layer 7 features it adds. The cost decision is therefore about what you actually need: if you only distribute TCP or UDP traffic across backends, a Standard Load Balancer is the right and cheaper tool, and you should reserve Application Gateway for cases that genuinely require path-based routing, TLS termination, or the managed web application firewall. Putting an Application Gateway in front of traffic that needs none of those features is a common and avoidable overspend.
How do you cut Azure routing costs, step by step?
These five steps right-size capacity, consolidate gateways, and clear reserved capacity you do not need.
- Right-size minimum capacity units and cap autoscaleSet the v2 minimum instance or capacity unit count to real baseline demand and set a maximum so a spike scales out without unbounded cost. The result is reserved capacity matched to demand with a known ceiling.
- Consolidate gateways behind shared listenersRoute multiple sites and paths through one gateway using host-based and path-based rules rather than one gateway per application. The result is one fixed hourly fee instead of many.
- Choose the cheapest SKU that meets the requirementUse Standard_v2 unless you need the managed WAF, and use a layer 4 Load Balancer where you do not need application routing at all. The result is the premium feature cost avoided wherever it is not needed.
- Move basic tiers off retired SKUsMigrate any Basic Load Balancer to Standard and any v1 gateway to v2, since the older SKUs are retired. The result is a supported, predictably billed configuration.
- Remove idle gateways, listeners, and public IPsDelete unused gateways and rules and release orphaned public IP addresses. The result is the silent fixed-fee waste eliminated.
Paying for a gateway per application?
Our Azure cost audit right-sizes capacity units, consolidates per-app gateways behind shared listeners, moves traffic to the cheapest SKU that fits, and clears idle gateways and public IPs. On the performance model you pay only from realized savings. No savings, no fee.
Book an Azure cost audit →Do I need WAF_v2 or is Standard_v2 enough?
Use Standard_v2 unless you specifically need the managed web application firewall, in which case WAF_v2 is required. WAF_v2 bills capacity units at a higher rate than Standard_v2, so the common mistake is applying the WAF SKU to every gateway by default, including internal services that face no untrusted traffic. Scope WAF_v2 to internet-facing applications that benefit from managed rule sets and bot protection, and keep internal or low-risk routing on Standard_v2. The same tier-matching discipline applies to the edge services in front of these gateways, covered in our guide to reducing Azure Front Door and CDN costs.
SKU names, billing models, and retirement dates reflect Azure as of June 2026. Verify current pricing and SKU availability on the linked Azure documentation before acting, because networking service packaging and rates change.
The Azure Cost Optimization Field Guide includes the gateway consolidation worksheet and the capacity-unit right-sizing checklist we apply on engagements. It is the downloadable companion to this article.
Frequently asked questions
How is Azure Application Gateway priced?
Azure Application Gateway v2 is billed in two parts: a fixed hourly fee for the time the gateway is provisioned and available, plus a capacity unit charge that scales with the compute, persistent connections, and throughput it consumes. The WAF_v2 SKU adds the managed web application firewall at a higher capacity unit rate than Standard_v2. Right-sizing the minimum capacity units and capping autoscale is the main cost lever. Verify current rates on the Azure Application Gateway pricing page.
What is the difference between Application Gateway and Load Balancer cost?
Azure Load Balancer is a layer 4 service billed by the number of load balancing rules and the volume of data processed, while Application Gateway is a layer 7 service billed by a fixed gateway-hour fee plus capacity units. Load Balancer is cheaper because it does no application-level routing or TLS termination, so use it when you only need to distribute traffic and reserve Application Gateway for path-based routing, TLS offload, or the web application firewall.
Do I need WAF_v2 or is Standard_v2 enough?
Use Standard_v2 unless you specifically need the managed web application firewall, in which case WAF_v2 is required. WAF_v2 carries a higher capacity unit rate, so applying it to every gateway by default inflates cost. Scope WAF_v2 to internet-facing applications that need managed rule protection and keep internal or low-risk traffic on Standard_v2.
How do I reduce idle Application Gateway cost?
Reduce idle cost by deleting gateways that no longer front live applications, lowering the minimum capacity unit count on lightly used gateways, consolidating several applications behind one gateway with host and path routing, and releasing orphaned public IP addresses. A provisioned v2 gateway bills its fixed hourly fee continuously, so an idle gateway is pure waste.
The short version
Azure routing savings come from sizing capacity units to real demand, consolidating per-application gateways behind shared listeners, choosing Load Balancer or Standard_v2 over WAF_v2 wherever it fits, and clearing idle gateways and public IPs. When you want your Azure networking estate right-sized and the per-app gateway sprawl collapsed, that is exactly what our Azure cost optimization service delivers.
Cloud pricing and service behavior change frequently. Verify the specifics in this guide against the providers’ own current documentation and the FinOps Foundation: Azure pricing ↗, Azure documentation ↗ and FinOps Foundation Framework ↗. This article also reflects Cloud Cost Room’s hands-on, vendor-neutral engagement experience.