# Resource Cleanup Service

*/Opportunities/Resource_Cleanup_Service*

## Opportunity Overview

**Wedge**: The beachhead targets orphaned AWS EBS volumes and unattached Elastic IPs. This narrow niche provides instant, mathematically provable ROI with virtually zero risk of impacting production traffic. Once trust is established through these safe deletions, the service expands into complex stateful resource deprecation, such as legacy RDS instances, and integrates into the CI/CD pipeline to prevent orphaned resources at deployment.
**Timing**: Extended context window models can now parse entire Terraform repositories alongside AWS CloudTrail logs to map complex resource dependencies accurately. Simultaneously, tighter capital environments force engineering organizations to prioritize cloud unit economics over unconstrained growth.
**Why This I C P**: Mid-market B2B SaaS platform engineering teams scale infrastructure rapidly but lack the dedicated FinOps headcount of large enterprises. They experience the acute pain of rising cloud bills without the internal bandwidth to manually prune their environments.
**Size Of Prize**: Approximately 100,000 mid-market to enterprise cloud-native organizations globally × ~$40,000 annual spend on manual FinOps labor and wasted cloud resources equals a ~$4B total addressable market.
**Gap Narrative**: Engineering teams pay for unused cloud infrastructure because identifying orphaned resources requires cross-referencing infrastructure-as-code state files with live cloud environments. Existing cost management tools generate static alerts but require human verification before deletion, leaving the actual cleanup work undone. Teams need a service that safely identifies, verifies, and executes the removal of dead assets without breaking production.
**Defensibility**: Defensibility stems from workflow lock-in and the creation of a proprietary organizational dependency graph. As the service maps how a specific engineering team tags, deploys, and deprecates resources, its confidence scores improve, making it difficult to replace with a generic scanner. Deep integration into daily Slack approval and GitHub PR workflows creates high switching costs.
**Why This Thesis**: A Service-as-Software model fits this problem because teams do not want another dashboard of recommendations; they want the labor of verification and deletion completed. Operating as an autonomous agent that submits pull requests and deletion runbooks directly replaces the human chore of infrastructure pruning.

## Opportunity Linked Thesis

**Thesis**: [Service-as-Software](/Theses/Service-as-Software)

## Opportunity Linked I C P

**Icp**: [Enterprise SaaS Provider](/CompanyTypes/Enterprise_SaaS_Provider)

## Opportunity Market Sizing

_Illustrative — target and order-of-magnitude estimate figures, not an achieved track record (this Thing is concept-stage)._

**S A M**: ~$400M-$600M US and EU enterprise SaaS providers
**S O M**: ~$10M-$25M
**T A M**: ~25k-30k global SaaS companies × ~$40k/yr ≈ ~$1B-$1.2B
**Growth Rate**: ~20-25%/yr, driven by exponential cloud resource sprawl and tightening FinOps mandates across SaaS engineering teams
**Paid Comparable Spend**: ~$60k-$120k/yr on legacy cloud cost management suites and dedicated SRE labor for manual cleanup scripts

## Opportunity Incumbents

- [AWS Trusted Advisor](/Products/AWS_Trusted_Advisor) — Tool
- [Cloud Custodian](/Products/Cloud_Custodian) — Open-Source
- [Resource Audit Spreadsheets](/Products/Resource_Audit_Spreadsheets) — Spreadsheet
- [Custom Python Scripts](/Products/Custom_Python_Scripts) — DIY
- [NetApp Spot](/Products/NetApp_Spot) — Tool
- [Accenture Cloud Optimization](/Products/Accenture_Cloud_Optimization) — Service

## Opportunity Win Conditions

**Kill Thresholds**:
- Fewer than 10% of flagged resources are actually deleted within 30 days
- Sales cycle exceeds 60 days for a $40k ACV contract
- More than 1 false-positive production deletion in the first 90 days
- D30 active usage drops below 20%
**Leading Metrics**:
- Time to first connected cloud account
- Percentage of identified idle resources deleted within 7 days
- Ratio of automated deletions to manual approvals
- Total dollar value of cloud spend reclaimed per week
**What Proves Right**: Engineering teams connect their cloud environments and authorize the automated deletion of at least 20% of flagged idle resources within the first 14 days. Customers convert to $3k/month paid contracts based entirely on the immediate hard-dollar savings demonstrated during the trial period. Cohort retention remains above 85% at month three as continuous automated cleanup integrates deeply into their infrastructure operations.
**What Proves Wrong**: Engineering teams install the service but refuse to enable auto-delete permissions, treating it as a read-only dashboard that is abandoned after the first billing cycle. Security and compliance reviews drag implementation past 60 days, skyrocketing customer acquisition costs. The system flags or deletes active production assets, immediately breaking trust and causing catastrophic churn.

## Opportunity Build Profile

**Hardest Part**: Safely identifying orphaned resources without causing production outages requires deep contextual awareness of stateful dependencies. Parsing fragmented cloud provider APIs to confidently distinguish between truly abandoned infrastructure and dormant disaster recovery systems is the make-or-break challenge.
**Min Viable Scope**: Target only AWS non-production environments, flagging just unattached EBS volumes and idle RDS instances. Deliberately leave out automated deletion, multi-cloud support, and complex Kubernetes workload rightsizing.
**Cold Start Problem**: Engineering teams refuse to grant destructive write permissions to an unproven startup tool. Break this by launching as a read-only analytics tool that generates manual Terraform or CLI cleanup scripts, earning trust through zero false positives before asking for automated execution rights.
**Time To First Value**: 1 hour to connect AWS read-only roles and generate the first actionable resource waste report.
**Data Moat Available**: false
**Technical Difficulty**: Moderate

## Neighborhood

### Where the gap lives

- [Declarative Provisioning API](/Agents/Declarative_Provisioning_API) — latent gap · Agents

### Incumbent in

- [Bespoke Python Scripts](/Products/Bespoke_Python_Scripts) — incumbent in · Products
- [AWS Trusted Advisor](/Products/AWS_Trusted_Advisor) — incumbent in · Products
- [Accenture Cloud Optimization](/Products/Accenture_Cloud_Optimization) — incumbent in · Products
- [Resource Audit Spreadsheets](/Products/Resource_Audit_Spreadsheets) — incumbent in · Products
- [Cloud Custodian](/Products/Cloud_Custodian) — incumbent in · Products
- [NetApp Spot](/Products/NetApp_Spot) — incumbent in · Products

### Applies thesis

- [Enterprise SaaS Provider](/CompanyTypes/Enterprise_SaaS_Provider) — applies thesis · CompanyTypes

### Embodies

- [Service-as-Software](/Theses/Service-as-Software) — embodies · Theses

### Similar Opportunities

- [FinOps Remediation Agent](/Metrics/Development_Cost_Per_Product/Processes/Engineering_And_Coding/Opportunities/FinOps_Remediation_Agent) — similar · Opportunities
- [Cloud Cost Remediation](/Opportunities/Cloud_Cost_Remediation) — similar · Opportunities
- [Fractional Systems Engineer](/Opportunities/Fractional_Systems_Engineer) — similar · Opportunities
- [Cloud FinOps Automation](/Opportunities/Cloud_FinOps_Automation) — similar · Opportunities
- [Cloud Provisioning Optimizer](/Occupations/Computer_and_Mathematical_Occupations/Opportunities/Cloud_Provisioning_Optimizer) — similar · Opportunities
- [Predictive Hibernation For DevOps](/Opportunities/Predictive_Hibernation_For_DevOps) — similar · Opportunities
- [FinOps Orchestration Engine](/Opportunities/FinOps_Orchestration_Engine) — similar · Opportunities
- [AI Systems Engineering](/Skills/Systems_Evaluation/Opportunities/AI_Systems_Engineering) — similar · Opportunities
- [Security Architecture Auditing](/Opportunities/Security_Architecture_Auditing) — similar · Opportunities
- [Automated Review for DevOps Teams](/Opportunities/Automated_Review_for_DevOps_Teams) — similar · Opportunities
- [Backlog Resolution Service](/Opportunities/Backlog_Resolution_Service) — similar · Opportunities
- [Capacity Tuning Engine](/Skills/Systems_Evaluation/Opportunities/Capacity_Tuning_Engine) — similar · Opportunities
- [AI Vendor Deduplication for Procurement](/Opportunities/AI_Vendor_Deduplication_for_Procurement) — similar · Opportunities
- [Automated Flag Cleanup For DevOps](/Opportunities/Automated_Flag_Cleanup_For_DevOps) — similar · Opportunities
- [Architecture Assessment Service](/Skills/Systems_Analysis/Opportunities/Architecture_Assessment_Service) — similar · Opportunities
- [Rogue Spend Interceptor](/Opportunities/Rogue_Spend_Interceptor) — similar · Opportunities
- [System Design Engine](/Opportunities/System_Design_Engine) — similar · Opportunities
- [Overhead Discovery Fabric](/Opportunities/Overhead_Discovery_Fabric) — similar · Opportunities
- [Access Policy Auditor](/Opportunities/Access_Policy_Auditor) — similar · Opportunities
- [Managed Auth Operations](/Opportunities/Managed_Auth_Operations) — similar · Opportunities