For platform teams running several clouds and LLM providers
Give every engineer the cost context of a FinOps team.
Costory joins your cloud, LLM and data-cloud bills with your telemetry, allocates every euro with rules you version in git, and answers where your engineers work: Claude, Slack, Backstage. From 250€/month (~$300). Never a % of your bill.
30-day free trial. Connect in under 30 minutes, get your first optimization recommendation the next day. Read-only on your cloud.
It starts in Slack, not in a dashboard.
One person curious about cost? Ask Claude and DIY. A whole org that needs the same number? Use Costory.
If one person in your company looks at the cloud bill, connect Claude to your billing API. It works, and you don’t need us. Once you start to slice your engineering team into multiple squads or tribes, and their leads and finance rely on the same number, it becomes a data engineering problem: one dataset, one allocation logic, one definition of “cost per customer”. That’s what Costory is.
Built by data engineers, on a modern data stack.
Every bill (AWS, GCP, Azure, Snowflake, Databricks, OpenAI, Anthropic and more) is normalized into one model on BigQuery and joined with your context: telemetry, CI, dbt, product events, business metrics. It’s also why we can charge a flat price: processing your bills costs us little, and we pass that on.
Four reasons to choose Costory
01
Allocation that holds in a large org.
Write allocation rules as YAML in git. Map your Backstage or CMDB catalog onto the bill, split shared costs by weight or by real usage, and date every entry so a reorg never rewrites last quarter. Reviewed in a PR, backfilled over your full history, managed with Terraform.
02
Everything included, at a flat price.
Unlimited users and sources. Notebooks, unit economics with formulas (cost per ticket, per customer, per feature), formula alerts you can backtest, budgets, Slack alerts and the MCP. In every plan, from 250€/month.
03
The FinOps harness.
An agent with the skills, the company context and the loops to monitor, investigate, report and recommend. Ask it in Claude, Cursor, Slack or Dust. Some teams only use Costory through Claude. Loops are Scouts: a markdown file that says what to read, what to summarize and when to ping. Costory runs it. It never touches your code base: it hands each saving to your coding agent.
04
Open by default.
Public API (OpenAPI). Export allocated data to your dashboards, your IDP (Backstage), BigQuery, Iceberg/Parquet or your finance systems. No lock-in.
Asked to build an internal FinOps tool?
One engineer with Claude and a billing API gets a useful dashboard in an afternoon. That part is free, and we recommend it. The cost starts when the whole team needs it: weeks to build a framework, tools and a harness others can share, then hosting, maintenance and every new feature request, forever, on the platform team’s backlog. We built that part. You keep control through YAML, formulas, the API and Terraform.
Clear limits
What Costory does not do
Costory does not buy Reserved Instances or Savings Plans for you. The agent recommends; your team decides.
It does not apply automated rightsizing or change resources for you.
It does not promise a savings percentage. Recommendations come with an impact estimate from your own usage.
It never gets write access to your cloud accounts. MCP actions create Costory objects only: alerts, reports, dashboards and events.
It never charges a percentage of your bill.
Start at 250€/month (~$300). Pay more only when you use more. Never a % of your bill.
FROM G2
What platform teams say on G2.
Read G2 reviews
“From 'How much do we pay' to 'How much each team is investing, in each tool', and why, at a glance”
“I appreciate the ability to manage budgets and track certain lines in detail. The MCP and its integration into my ecosystem are also a real plus.”
“Fast, Smooth MCP Server Integration with Responsive Support”
