Azure App Service Plan tiers cost: you pay for the plan, not the app
App Service Plans bill per instance-hour by tier (B1 ~$13, S1 ~$70, P1v3 ~$120+/mo) times instances. Apps share a plan for free. Here's how to pick the tier and why consolidating apps is the big saving.
Quick answer
An App Service Plan bills per instance-hour at its tier rate, continuously, times the instance count: Basic B1 ~$13/mo, Standard S1 ~$70/mo, Premium P1v3 ~$120+/mo. Multiple apps share one plan at no extra compute charge — so the saving is consolidating apps onto shared plans, right-sizing the tier to features you use, and using autoscale instead of a fixed peak instance count.
The App Service Plan is the compute you pay for; the web apps, function apps, and APIs that run on it are free to add. That one fact drives most App Service cost decisions — and the most common overspend is creating a separate plan per app instead of sharing.
The tier ladder
| Tier | Example | ~Cost | What it adds |
|---|---|---|---|
| Free / Shared | F1 / D1 | $0–low | Experiments; no custom SSL on Free, quotas |
| Basic | B1 | ~$13/mo | Custom domains, manual scale, dev/test |
| Standard | S1 | ~$70/mo | Autoscale, staging slots, daily backups |
| Premium v3 | P1v3 | ~$120+/mo | More CPU/RAM, VNet, better scale, reservable |
| Isolated | I1v2 | highest | Dedicated/compliance (App Service Environment) |
Rates are per instance; scaling to 3 instances triples the line. Pick the tier by the lowest one that has the feature you need — paying Premium for a workload that only needs Standard's slots and autoscale is pure overspend.
Consolidate apps onto shared plans
Because apps share a plan's compute for free, ten low-traffic apps on one S1 plan cost $70/month total — not $700. The anti-pattern is a plan per app "for isolation," which multiplies the per-instance fee for capacity that sits mostly idle. Group low-traffic apps; give high-traffic ones their own plan so they don't contend.
Autoscale, don't pin to peak
Standard and Premium support autoscale rules. A plan fixed at the instance count you need for peak pays for peak 24/7. Autoscaling out under load and back down when idle pays for peak only when it happens — the same logic as Consumption vs Premium Functions and AKS node pools.
Reserve and schedule
- Reserve Premium v3 for 1-3 years on steady production plans — 30-55% off.
- Stop or scale down dev/test plans off-hours. A B1 dev plan running only business hours costs a third of a 24/7 one.
FAQ
How is an Azure App Service Plan priced?
Per instance-hour at the plan's tier rate, billed continuously regardless of traffic, and multiplied by the number of instances you scale to. Free and Shared tiers are $0 or near it (with limits); Basic B1 is ~$13/month; Standard S1 ~$70/month; Premium P1v3 ~$120+/month. You pay for the plan, and any number of apps can run on it.
Which App Service tier should I use?
Free/Shared for experiments only (no custom domains/SSL on Free, sleep limits). Basic for dev/test with custom domains. Standard for production needing autoscale, staging slots, and daily backups. Premium v3 for production needing more performance, VNet integration, and better scale. Isolated for compliance/dedicated environments. Match the tier to the feature you actually need.
Does each app cost extra on an App Service Plan?
No — you pay for the plan (the compute), and multiple apps can share one plan at no additional compute charge. The trap is the opposite: creating a separate plan per app multiplies the per-instance cost. Consolidating low-traffic apps onto a shared plan is a common saving.
How does scaling affect App Service cost?
Linearly. Each instance you scale out to bills at the full tier rate. Autoscaling (Standard tier and up) adds instances under load and removes them when idle, so you pay for peak only when peak happens. A plan pinned at a fixed high instance count pays for peak 24/7.
How do I reduce App Service costs?
Consolidate apps onto shared plans, right-size the tier to the features you use (don't pay Premium for Standard needs), enable autoscale instead of fixed instance counts, scale dev/test plans down or stop them off-hours, and reserve Premium v3 instances (1-3 year) for steady production.
How does C3X estimate App Service cost?
C3X prices an azurerm_service_plan from its tier and instance count, so the per-instance monthly cost and the effect of scaling out are visible before deployment.
What to do next
The plan tier and instance count are in your Terraform, so the monthly cost is decidable up front. C3X prices an azurerm_app_service_plan from its tier and instances, so an oversized tier or a per-app plan sprawl shows up in review. The quickstart runs it in minutes.
Share this post
Try C3X on your own Terraform
Free and open source. No API key required. One command to install, one command to estimate.