Make cloud cost everyone's normal business
Tagging, cost allocation, showback, budgets, anomaly alerts and unit economics, set up so cost becomes a normal part of how engineering works.
Overview
A one-off cost cleanup helps for a few months. Then the bill drifts up again, because nothing changed in how teams make decisions. A FinOps operating model gives engineering, finance and product the same view of spend and a routine for acting on it.
We put the plumbing in place first: a tagging standard, enforced in code, and cost data split by team and product. Then we add the habits: a monthly review, budgets per team, anomaly alerts and a unit cost such as cost per customer or per transaction that leadership can track.
Signs you need this
- Finance and engineering disagree on what the cloud bill pays for
- Most resources have no owner or cost centre tag
- Unexpected spend is found weeks later on the invoice
- No one can state what it costs to serve one customer
What you get
- Tagging standard and enforcement
- A short list of required tags, applied through infrastructure code and checked by policy so untagged resources are flagged or blocked.
- Cost allocation model
- Rules for splitting direct and shared costs, such as networking and support, across teams and products.
- Showback reports
- Monthly cost views per team and product, delivered where people already look, such as a dashboard or a Slack channel.
- Budgets and anomaly alerts
- Per-team budgets and anomaly detection in your cloud provider, routed to the people who can act.
- Unit cost metrics
- One or two unit costs, such as cost per customer or per thousand transactions, joined from billing and product data.
- Review routine
- A monthly meeting format with agenda, inputs and owners, run with you for the first cycles.
How the work runs
- 1
Stakeholder alignment
1 week
Agree with finance, engineering and product what questions cost data must answer and who owns what.
- 2
Tagging and allocation
2 to 4 weeks
Define the tagging standard, backfill tags on existing resources and build the allocation rules.
- 3
Reporting and alerts
2 to 3 weeks
Build showback reports, unit cost metrics, budgets and anomaly alerts from your billing exports.
- 4
First review cycles
4 to 8 weeks
Run the monthly review with your teams, refine the reports and hand over ownership.
Timings depend on the size of your estate. The assessment fixes them before any build starts.
What you can check afterwards
- Each team receives a monthly cost report for its own services
- New resources without required tags are flagged or blocked by policy
- Anomaly alerts reach a named owner within a day of a spending jump
- Leadership can track an agreed unit cost from month to month
Tools we work with
- AWS Cost and Usage Report
- AWS Cost Anomaly Detection
- AWS Budgets
- Azure Cost Management
- Google Cloud Billing export
- Amazon Athena
- BigQuery
- Grafana
- Open Policy Agent
- Terraform
- FOCUS billing specification
Questions buyers ask
- Do we need a paid FinOps platform?
- Not to start. Provider billing exports, a query engine and Grafana cover most needs. We only suggest a paid tool once you have outgrown them.
- How much time does this take from our teams?
- Some at the start for tagging, then about an hour a month per team lead for the review. The aim is a light routine people actually keep.
- What access do you need?
- Billing data access and read access to resource configuration in your own accounts, plus time with someone from finance.
- Is this the same as cloud cost reduction?
- No. Cost reduction is a project that removes waste now. This builds the ongoing process that stops waste returning, and the two work well together.
Talk to us about FinOps operating model.
Tell us what you run and where. We reply within one business day with questions or a quote.