1
Define the policy
PUT /tag-intelligence/policy with the required keys (e.g. env, owner, cost-center) and allowed values per key. The Tag Score is the percentage of required tags present across the tenant's resources.
required keys + allowed values
Tag Intelligence
An automated tagging governance layer that surfaces untagged spend, suggests likely owners using ML, detects drift from your tag policy and produces remediation tasks ready for IaC.
- ML owner suggestions
- Tag policy enforcement
- Terraform drift detection
lumicost.com — tag-intelligence
By the numbers
Tag Intelligencein 4 numbers
missing · inconsistent · duplicate · misassigned · IaC drift
0
tag issue rules
presentTags / requiredTags × 100
0
max tag score
GET / PUT /policy · required keys + allowed values
0
editable tag policy
TAG_SCORE_DROP
0
score-drop alert type
How it works
Three steps. No agents. No surprises.
2
Surface issues
Five rules detect MISSING_TAG, INCONSISTENT_VALUE, DUPLICATE_TAG, MISASSIGNED_COST and IAC_TAG_DRIFT. Each issue is linked to its resources, severity and a $/mo cost impact.
5 rules · with $/mo
3
Suggest & resolve
POST /suggest returns canonical values for inconsistent tags. Resolving is a single call to /issues/{id}/resolve. Score drops fire a TAG_SCORE_DROP alert through the alerts engine.
suggest · resolve · alert
What's inside
An automated tagging governance layer that surfaces untagged spend, suggests likely owners using ML, detects drift from your tag policy and produces remediation tasks ready for IaC.
01
Stop hunting for owners in Slack
We correlate naming patterns, IAM activity, deployment history and adjacent resources to suggest the most likely owner of any untagged resource — accept with one click and the tag is propagated.
02
Policy as code, drift as data
Declare your tag policy (required keys, allowed values, regex constraints) once. Drift is reported continuously, with the offending resources joined to the cost they're already burning.
03
Remediation that lands as a PR
Bulk-fix proposals are generated as Terraform / OpenTofu diffs against your repo — review, approve, merge.
Capabilities
Everything you need on day one.
- MISSING_TAG detection
- INCONSISTENT_VALUE normalisation
- DUPLICATE_TAG detection
- MISASSIGNED_COST flagging
- IAC_TAG_DRIFT detection
- 0–100 tag score per tenant
- Editable policy: required keys + allowed values
- TAG_SCORE_DROP alert type
Before vs after Lumicost
Night and day.
Without Lumicost
- 'env: Prod' vs 'env: production' chaos
- Untagged spend is invisible
- You only notice tag rot at the next audit
With Lumicost
- INCONSISTENT_VALUE flagged with canonical fix
- MISSING_TAG impact shown in $/mo
- TAG_SCORE_DROP alert when it slips
Where teams use it
Three ways teams put it to work.
Onboarding a new BU
Map an entire business unit's footprint to teams in 1–2 days using ML suggestions instead of weeks of forensics.
Compliance evidence
Prove to auditors that 100% of in-scope production resources carry the required ownership tags.
Cost allocation accuracy
Lift unallocated cost from 22% to under 3% in a quarter — and keep it there.
From the field
“A tag score is only useful if dropping it sends an alert. We made TAG_SCORE_DROP a first-class alert type for that reason.”
Plays nice with your stack
AWSGCPAzureOracle CloudKubernetesTerraformPulumiSlackJiraLinearGitHubGitLabDatadogSplunkOpenTelemetryPagerDutyNetSuiteSAPAWSGCPAzureOracle CloudKubernetesTerraformPulumiSlackJiraLinearGitHubGitLabDatadogSplunkOpenTelemetryPagerDutyNetSuiteSAP
FAQ
Frequently asked questions
Will Lumicost change tags directly?
Only if you ask via the optional write-mode adapter; by default we emit IaC diffs and let your pipeline apply them.
How accurate is the owner ML?
Internally measured at >88% top-1 accuracy on customers with ≥6 months of CloudTrail / Audit Log history.
Keep exploring
Try it free.
Connect read-only credentials. First insights in 24 hours. No credit card.
90 seconds · read-only credentials · no credit card