Azure Application Insights pricing: ingestion volume, sampling, and retention

Application Insights is a telemetry pipeline. Costs are predictable if you model it as: events * size -> GB ingested -> retention storage -> query scans.

Quick Application Insights estimate

  • Ingestion GB/day: trace + exception + dependency payload size.
  • Retention days: longer retention multiplies storage baseline.
  • Sampling rate: client/server sampling changes billable GB.

0) Define what you send

  • Telemetry types: requests, dependencies, exceptions, traces, custom events.
  • High-volume paths: a few endpoints often generate most traffic (and most telemetry).
  • Baseline vs incident: incidents multiply errors, retries, and event sizes.

1) Ingestion volume (GB/month)

A reliable first estimate is: requests * events per request * bytes per event. Model separate lines for high-volume sources rather than one blended average.

Tool: Log/telemetry ingestion.

2) Sampling (the biggest cost lever)

Sampling is the highest leverage knob. If you sample traces at 10%, ingestion can drop by ~10x for that stream. During incidents, exceptions and retries can multiply events, so sampling and filters matter.

  • Prefer targeted full-fidelity for short windows (investigation) over always-on full-fidelity.
  • Keep an incident plan: what gets sampled more or less when traffic spikes.

3) Retention and query scans

Retention multiplies stored GB. Dashboards and alerts can scan large windows repeatedly; model refresh frequency and scan windows explicitly so the "analysis bill" does not surprise you.

Tools: Retention storage, Scan/query.

Worked estimate template (copy/paste)

  • Ingestion GB/month = requests/month * events/request * bytes/event / 1e9 (approx)
  • Retention GB-month = ingestion GB/day * retention days (order-of-magnitude)
  • Query scans = dashboard refreshes/day * scan window (days) * ingestion GB/day

Common pitfalls

  • Using one average: a few endpoints can dominate traffic and telemetry.
  • Leaving traces at full fidelity for all services and all environments.
  • Not accounting for incident multipliers (errors and retries can 10x event volume).
  • Dashboards scanning wide windows frequently (quiet scan cost).
  • Keeping long retention by default without a clear reason.

How to validate the estimate

  • Validate sampling and filters and confirm what is kept vs dropped.
  • Validate incident behavior: errors and retries multiply telemetry events and ingestion volume.
  • Validate dashboards/alerts for wide time windows that scan lots of data frequently.

Related tools

Log cost tools Ingestion Retention Scan

Sources

Related guides

Azure Monitor metrics pricing: estimate custom metrics, retention, and API calls
A practical metrics cost model: time series count, sample rate, retention, and dashboard/alert query behavior. Includes validation steps to avoid high-cardinality mistakes.
Azure Log Analytics pricing: ingestion, retention, and query costs
A practical model for Log Analytics-style costs: GB ingested, retention storage, and query/scan behavior. Includes a method to estimate log GB from event rate and payload size, plus a validation checklist for high-volume sources.
Azure Event Hubs Pricing & Cost Guide (throughput, retention, egress)
Estimate Event Hubs cost from throughput units, ingress, capture/storage, and egress. Includes a planning checklist.
Azure SQL Database pricing: a practical estimate (compute, storage, backups, transfer)
Model Azure SQL Database cost without memorizing price tables: compute baseline (vCore/DTU), storage GB-month + growth, backup retention, and network transfer. Includes a validation checklist and common sizing traps.
CloudWatch Logs pricing: ingestion, retention, and queries
A practical CloudWatch Logs pricing guide: model ingestion (GB/day), retention (GB-month), and query/scan costs (Insights/Athena). Includes pitfalls and a validation checklist.
Azure API Management pricing: model requests, transfer, and log volume
A practical API Management estimate: request volume, response transfer, and logs/observability. Includes a checklist to validate retries, payload size, and usage tiers.

Related calculators

Metrics Time Series Cost Calculator
Estimate monthly metrics cost from active series and $ per series-month pricing.
CloudWatch Metrics Cost Calculator
Estimate CloudWatch metrics cost from custom metrics, alarms, dashboards, and API requests.
AWS CloudWatch Alarms Cost Calculator
Estimate alarm-month charges from standard, high-resolution, and composite alarm counts.

FAQ

What usually drives Application Insights cost?
Telemetry ingestion volume (GB) is usually the main driver. Retention and query volume become meaningful when you keep data longer or run frequent analytics.
How do I estimate quickly?
Estimate events per request, average event size, and request volume. Convert to GB/day and GB/month, then apply retention days and query scan behavior.
How do I validate?
Validate sampling rates, validate incident behavior (errors and retries multiply events), and validate retention settings for each telemetry type.
Last updated: 2026-02-07