Skip to main content
Your AWS bill spiked 30% last week. Was it that deploy on Tuesday, or something else? overlays deploys, purchases, and custom events directly on your cost charts so you can answer that question in seconds. Custom metrics let you go further: divide cost by usage to track efficiency over time, not just total spend.
QuestionCapabilityExample
What changed?Cost Explorer”EC2 spend increased 30% last week”
Why did it change?Event correlation”A deploy on March 5th added a new batch job”
Does it matter?”Cost per active user still dropped 12%“
Who should know?Digest / Reports”AI summary sent to #cloud-costs in Slack”

​
Event correlation

Overlay technical and business events directly on your cost charts to spot the cause of cost changes at a glance, without switching between your billing console, CI/CD tool, and Slack threads.
Cost chart where deploy and purchase event markers align with cost changes, showing cause and effect at a glance

​
When to use event correlation

  • You see a sudden increase in EC2 spend that lines up with a deploy marker. Click the event to see which PR was merged and what it changed.
  • A purchase event appears on the chart, letting you verify the cost line drops as expected in the following days.
  • Attach event overlays to a saved view so colleagues see why costs changed, not just that they changed.
  • During an outage or cost anomaly, event markers help you quickly rule in or rule out recent deploys or infrastructure changes.

​
Supported event types

Costory groups events into categories you can toggle independently:
  • Cloud provider events: purchases, CUD activations, marketplace renewals, detected automatically from your billing data
  • Technical events: GitHub pull requests merged, production deploys, CI/CD pipeline events
  • Business events: custom events you push via the public API, such as feature launches, marketing campaigns, pricing changes, or anything relevant to cost
Events panel with toggleable categories (Business, Technical, Commitments, Marketplace), each with Add to chart buttons

​
How to use event correlation

1

Open Cost Explorer

Navigate to Cost Explorer and build your query.
2

Enable event overlays

Click Show Events in the toolbar. The Events panel opens, showing all available events grouped by category. Toggle categories on or off, or add individual events to the chart.
Toggling GitHub deploy events on and Savings Plan events off to focus on a specific cost spike investigation
3

Investigate a cost change

Event markers appear as numbered badges on the chart timeline. Hover over any marker to see the event type, timestamp, and description. Correlate the timing of the event with the cost change to identify the root cause.
Cost chart where a numbered deploy marker on March 5th aligns with a 30% cost increase, hover reveals PR details
Combine event overlays with the Cost Waterfall view: the waterfall shows which services drove the cost change, and the events show why.

​
Example: investigating a cost spike after a deploy

Imagine your weekly EC2 bill jumped 30% starting March 5th. Here’s how event correlation helps:
  1. Open Cost Explorer and filter to EC2 costs for the past two weeks. You see the spike clearly in the chart.
  2. Enable GitHub events. A deploy marker appears on March 5th, the same day the spike started.
  3. Click the event to see that PR #1234 was merged: “Add nightly batch processing job for ML model retraining.”
  4. Root cause identified: the new batch job runs large GPU instances every night. You can now decide whether to optimize instance types, adjust scheduling, or accept the cost.
Without event correlation, you’d spend 30 minutes cross-referencing deployment logs, Slack threads, and billing data. With it, the answer is on the chart.

​
Attach events to saved views

You can attach event overlays to saved views so the context travels with the data when you share it with your team. When a colleague opens the view or receives it as a scheduled report, they see both the cost data and the events that explain it.
For automated cost change detection, the Digest surfaces the highest-impact changes and generates AI summaries you can share directly, no manual investigation needed.

​
Managing events

The Events page gives you a central view of all events across your organization. You can search, filter by source or tag, and manually create events for milestones that don’t come from an integration (e.g., “Q1 pricing change” or “data center migration started”).
Central Events page showing deploy events, Savings Plan purchases, and custom events, filterable by source and tag

​
Setting up event sources

To overlay events in Cost Explorer, you need at least one event source connected:

GitHub

Deploys and PR merges

CI/CD

Pipeline events

Custom API

Push any event
Cloud provider events ( purchases, CUD activations) are detected automatically from your billing data. No additional setup required.
Pushing a custom event is a single POST /v1/events call with a JSON body. Drop it into your CI/CD pipeline in under 5 minutes. See the Custom API guide for the full spec.

​
Unit economics and custom metrics

Define formulas to track cost efficiency alongside raw spend. Instead of only seeing “we spent $50k on compute,” you can answer “our cost per daily active user dropped 12% this month.”
Total cloud spend up 63% while cost per client down 26%, proving infrastructure investment is scaling efficiently

​
What you can build

  • Track infrastructure unit economics like cost per GB transferred, cost per build minute, or cost per container-hour by dividing cloud spend by usage metrics from your billing data or from sources like Datadog.
  • Track application unit economics like cost per API call, cost per DAU, or cost per transaction by connecting external metrics from Amplitude, BigQuery, or CSV. See the Cost Per Active User guide for a step-by-step example.
  • Compare the cost of a new service against the one it replaced by grouping formulas by service tag and tracking the trend over time.
  • Track realized savings from or CUDs by computing list_cost - total_cost.
  • Build custom KPIs for Savings Plan and CUD utilization rates.
You can apply these patterns to concrete scenarios like BigQuery cost visibility with dbt or allocating shared database costs.

​
Available metric types

Costory’s query bar supports four types of metrics you can combine in formulas:
Built-in billing metrics from your connected cloud providers. These are always available and include fields like total cost, , list cost, and usage quantity.

​
Prerequisites

To build unit economics formulas, you need at least one usage metric source connected to Costory. Supported sources include:
You don’t need external metrics to use formulas. You can combine cost fields with usage metrics from your billing data (e.g., cost per GB-Hour) without connecting any external source.

​
Creating a custom formula

1

Open Cost Explorer

Navigate to Cost Explorer and configure your base cost query.
2

Add metrics

Click + Add Metric in the query bar to add a cost metric or an external metric. Each metric gets a letter label (a, b, c…) that you reference in formulas.
3

Add a formula

Click + Add Formula. Write your formula using the letter labels. For example:
  • If metric a is total cost and metric b is GB-Hours: a / b gives you cost per GB-Hour
  • If metric a is total cost and metric b is Daily Active Users: a / b gives you cost per DAU
Formula editor combining total cost (a) divided by usage metric (b) to produce cost-per-unit over time
Formulas currently support +, -, *, /, and parentheses. For more complex transformations (rolling averages, conditionals), pre-compute the metric in your source system before importing.
4

Visualize the result

The chart updates to show your formula result over time. You can switch between chart types (bar, line) and toggle between the raw metrics and the formula output.
Cost per DAU trending down over time even as total cloud spend increases, dual-axis chart showing both metrics
5

Save and share

Save the query so you can revisit it, export it to a dashboard, or schedule it as a report.
Formulas respect your group-by dimensions. If you group by team and define cost / daily_active_users, you’ll get a separate cost-per-DAU value for each team, useful for comparing efficiency across teams.
You can set threshold alerts on any formula. If cost-per-request exceeds your target, Costory sends a notification via Slack or email. Configure alerts in the Alerts settings.

​
Frequently Asked Questions

Connect your usage metrics (e.g., DAU from Amplitude, transactions from your database) and create a formula that divides cloud spend by your usage metric. See the Cost Per Active User guide for a step-by-step example. You can apply the same approach to BigQuery costs with dbt or shared database costs.
Yes. Costory integrates with GitHub to overlay deploy events on your cost charts. You can also push custom events via the API. This makes it easy to explain cost spikes to your team without leaving the Cost Explorer.
Yes. Events like Savings Plan purchases, CUD activations, and marketplace renewals are extracted from your billing data automatically. No additional setup is needed beyond connecting your cloud provider.
Formulas support arithmetic operators (+, -, *, /) and parentheses for grouping. Each metric is referenced by its letter label (a, b, c…). Currently, formulas operate on the aggregated value per time bucket. There are no built-in functions like min, max, or percentile, and no conditional logic. If you need rolling averages or more complex transformations, pre-aggregate in your external metric source (e.g., Datadog or BigQuery) before importing into Costory.
Yes. If you group by a dimension (e.g., team or service), the formula is applied per group. This lets you compare unit economics across teams, services, or any other dimension.
Costory processes your cloud billing data daily, typically within 24 hours of your cloud provider publishing it. You can query as far back as your billing export history goes. If you configured your CUR or billing export 18 months ago, you have 18 months of data. Multi-account setups and consolidated billing are fully supported. For details on connecting your cloud provider, see the setup guides.
CapabilityAWS Console / CloudWatchGrafana + Cost pluginsCostory
Actual billing data (CUR/export)✅ (Cost Explorer)⚠️ Requires custom ETL✅ Native
Deploy / PR event overlay❌⚠️ Manual annotations✅ Automatic from GitHub
Cost allocation by team/service⚠️ Tags only❌✅ Tags + virtual dimensions
Unit economics (cost ÷ business metric)❌⚠️ Custom queries✅ Built-in formula editor
Shareable saved views with context❌⚠️ Dashboard links✅ One-click share with events attached
Costory doesn’t replace your observability stack. It fills the gap between “something is expensive” and “this is why, and this is who should fix it.”

​
Next Steps

Cost Explorer

Build the base queries for your analysis

Quick Share

Save views, export to dashboards, and share with your team

Cost Per Active User

Step-by-step guide to tracking cost per DAU with Amplitude

Unit Economics Use Case

Advanced formulas and reporting workflows

Event Correlation Use Case

Full walkthrough of using events to explain cost changes

Digest

Auto-triage cost anomalies with AI-generated summaries
Last modified on March 18, 2026