- Industry
- Retail
- Department
- Engineering
- Team
- Platform and Reliability
- Kind
- Decision-maker
- Approver
- Head of Platform
- Code
- ENG-13
It finds wasted cloud spend across the online store, commerce services, data platform and test environments, such as capacity left scaled up after a sale event or idle test stores, and decides what to resize, schedule or remove. It forecasts the saving and the risk to trading for each.
How it works
7 steps, and a person signs off before step 7.
- 01Looks it up
Finds the record the request is about in the cloud billing and usage data: spend, utilisation and tags by resource for the scope and period..
- 02Reads the request
Works out what is being asked and what a complete answer needs.
- 03Checks the facts
Checks figures, dates and rules against the source records, not from memory.
- 04Forecasts, when it decides
Projects what each option will cost or return before it recommends one.
- 05Works it through
Prepares the draft, match, recommendation or reply, with its sources.
- 06Asks a person
Hands the work to Head of Platform with its reasoning before anything goes out.
- 07Acts
Once approved, it acts in the code repository: it opens the infrastructure-as-code changes for the approved cost actions. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The cloud cost guide: spend by service and team, the reserved capacity commitments, the sale event scaling calendar, the environments allowed to sleep, and the cost tags.
Reads from
the cloud billing and usage data: spend, utilisation and tags by resource for the scope and period.
Acts in
the code repository: it opens the infrastructure-as-code changes for the approved cost actions.
Hands back
A list with one entry per action, each with resource, service, action (resize, schedule, remove or commit), saving_per_month, trading_risk (none, low or high) and reason.
Asks a person when
A change touches production, a reserved capacity commitment, or any resource in the card data environment.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| CPU below which a resource is idle (%) | 5 | A resource averaging below this for two weeks is idle. |
| Smallest saving worth acting on (USD a month) | 200 | Smaller savings are listed but not actioned. |
| Sleep test environments out of hours | On | When on, non-production environments sleep at night and weekends. |
| Scale down after a sale event (days) | 3 | Capacity raised for a sale event is reduced this many days after it ends. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Never resize or remove production capacity within two weeks of a sale event or during the freeze.
- Scale down capacity raised for a sale event 3 days after it ends.
- A resource averaging below 5 percent CPU for two weeks is idle; propose removing or resizing it.
- When on is true, schedule test environments to sleep out of hours.
- Act only on savings of 200 or more a month; forecast the risk to trading for each.
See the Cloud Cost Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.