New Savings Proposals: approve, test and roll back cost changesLearn more Sign in|Talk to a cloud engineer
Product tourGet a demo

Map the bill to teams and products

Write rules that turn accounts, services and tags into the names your business uses, and see the dollar impact of every rule before you publish it.

Get a demoTry the interactive demo
What you're watching
  1. 1
    Add rule 4 to Product line

    Three rules already map teams and Bedrock to Storefront, Data & Search and AI Assistant. The new rule sends account sandbox to Experiments.

  2. 2
    Check the preview

    Allocated spend rises from 82.2% to 94.2% and Experiments appears as a new row. The Legacy rule is flagged as never firing.

  3. 3
    Review the diff and publish v8

    The v7 to v8 diff shows one added rule and unallocated falling from $32,808 to $10,690. Publishing puts it in Cost Explorer.

Who does thisFinOps lead, with finance reviewing the diffWhat you get94.2% of spend mapped to product lines, with the change saved as version 8.
2dimension types: grouping and allocation
1stmatching rule wins, evaluated top to bottom
$shown for every rule in the preview
Allpublishes kept as versions with diffs

Shared costs end up in nobody's budget

Kubernetes, data transfer and support fees rarely carry a team tag. Allocation rules decide where they land.

Ownership

Tags that don’t cover everything

What usually happens: Untagged resources and shared accounts leave a large “other” bucket nobody owns, and it grows every quarter.

How CloudLens resolves it: Rules match on account, service, region, usage type and tags together. Whatever still doesn’t match shows as Unallocated.

Resolved
Allocation

Shared platform cost has no owner

What usually happens: The Kubernetes cluster and the data lake live in a shared account, so one team carries the cost or it gets split evenly by guesswork.

How CloudLens resolves it: An allocation dimension spreads shared cost across the teams that use it, shown next to each team’s direct spend.

Resolved
Audit

Rule changes rewrite last quarter

What usually happens: Someone edits the mapping sheet and last quarter’s numbers quietly change.

How CloudLens resolves it: Every publish is a version. Compare two versions line by line and restore an older one as a new version.

Resolved

How Ana got Checkout’s real cost before budget season

Two hours of rule-writing that turned a mostly unallocated bill into numbers each team could defend.

ARAna RuizFinOps lead, Lumora Retail

Lumora Retail is a fictional company. The people, names and numbers are sample data.

    1
    Tue 10:0011% unallocated

    Starts from the Unallocated line

    Cost Explorer grouped by Team shows 11% of last month as Unallocated, mostly in prod-core and prod-data.
    2
    Tue 10:15

    Writes a rule in the condition builder

    Account is prod-core and tag service starts with “checkout-” becomes Checkout. The preview shows the rule would claim $38,204 of last month.
    3
    Tue 10:25

    Catches a rule that never fires

    The preview marks an older checkout-legacy rule as never firing, because the new rule above it matches the same rows first. She deletes it.
    4
    Tue 10:40

    Allocates the shared EKS cluster

    An allocation dimension spreads platform-eks across the teams that run workloads on it, so Checkout carries its share instead of Shared Services.
    5
    Tue 11:05v14 published

    Publishes version 14

    The diff shows two changed rules. Unallocated drops to 3%, and the new Team grouping is live in Cost Explorer for everyone.
    6
    Thu 16:00

    Finance asks what changed

    Version history shows who published v14, when, and the exact lines that changed. The answer is a link, not a meeting.
Still arguing about who owns shared cost?

Map your teams with our engineers

Get a demo

Rules you can read, results you can check

Build conditions with the builder or the code editor. The preview shows what each rule matches, which rules never fire, and how much is still unallocated.

Give spend one name, or split it fairly

Grouping gives every row of spend exactly one name. Allocation spreads shared cost, such as a platform cluster or a data lake, across the teams that use it. In the example, a $22,117/mo EKS platform cluster is split by CPU-hours per namespace.
  • Grouping: one name per row
  • Allocation: shared cost spread by a driver
  • Direct and allocated side by side
What you're watching
  1. 1
    Start in Grouping

    Each row of spend gets one name, so the four team cards show direct spend only.

  2. 2
    Switch to Allocation

    The shared EKS platform cluster, $22,117/mo, is split by CPU-hours per namespace instead of landing on one team.

  3. 3
    See each team's share land

    Checkout takes 34%, or $7,520. Search & Discovery, Data Platform and Mobile API pick up the rest, shown next to direct spend.

Who does thisFinOps lead, with the platform team that runs the clusterWhat you getEach team's number includes its share of the shared cluster, with the split visible.

Every publish is a version you can compare

See who changed which rule and when, compare any two versions line by line, and restore an older one. A restore creates a new version, so history is never rewritten.
  • Line-by-line diff
  • Restore creates a new version
  • Author and time on every change
What you're watching
  1. 1
    Open version history

    Product line lists v5 to v8, each with its author, date and note, such as Ravi's “Split Data & Search” on Sep 3.

  2. 2
    Compare v7 with v8

    The diff removes the rule mapping Data Platform to Data and adds one mapping two teams to Data & Search.

  3. 3
    Restore v7 as v9

    Restore v7 saves the older rules as a new version 9. Version 8 stays in history.

Who does thisFinOps lead, or finance checking how last quarter was mappedWhat you getAn older mapping back in place, with every earlier version still on record.

Published dimensions show up in Cost Explorer

See Cost Explorer
Once a dimension is published it appears under “Your Dimensions” in every group-by and filter, so a product owner can read their own spend without learning your account structure.
  • Group, filter and drill by it
  • Build saved views on it
  • Unallocated stays visible
What you're watching
  1. 1
    Group by Service, then Product line

    The chart restacks into Storefront, Data & Search, AI Assistant, Experiments, Platform and Unallocated.

  2. 2
    Change the chart and compare

    Line and stacked area show the same total. Compare to Month Ago overlays the previous 30 days.

  3. 3
    Click a bar to drill in

    Sep 5 opens split by service, with Group by Usage type offered as the next level down.

Who does thisProduct owner or engineering managerWhat you getProduct owners read their own spend in the names they already use.

Allocation you can defend in a budget review

First match wins

Rules are evaluated top to bottom, so it is always clear which rule claimed a dollar.

Preview before publish

Dollars per rule, rules that never fire and rules shadowed by an earlier one, all visible before anything changes.

History you can audit

Versions, diffs and restores mean finance can always see what the rules were last quarter.

Cost Explorer Budgets per team

Dimension Studio questions

Any billing dimension, such as account, service, region or usage type, plus your tags. Conditions can be combined, and each rule outputs one value.

It shows as Unallocated, in the preview and in Cost Explorer, so gaps stay visible instead of being absorbed somewhere.

Create an allocation dimension and choose the shared cost. It is spread across the owners that use it and shown next to their direct spend.

Yes. Open version history, compare versions and restore the one you want. The restore is saved as a new version.

Ready to see CloudLens in action?

Connect a read-only AWS role or Azure service principal. We'll walk you through your bill, your security graph and the first things worth fixing.