When original engineers leave, internal tools rot silently — patched by contractors with no context — until they become operational nightmares that nobody wants to own.
Integrates with git, HR/org data, and incident tools to detect 'orphaned' services (no active maintainer, high contractor commit ratio, rising incident count, stale dependencies) and surfaces them in a dashboard with ownership assignment workflows.
SaaS subscription ($1000-$3000/mo based on org size)
The Reddit thread (79 upvotes, 91 comments) validates visceral pain. Every engineer at a 500+ person company has a horror story about an orphaned service. The pain is acute when it manifests (production incidents from unmaintained services) but chronic and invisible beforehand — which is exactly why a detection tool is needed. Deducting points because the pain is episodic, not daily, and many orgs tolerate it until it explodes.
Target is platform engineering leaders at 200+ engineer orgs. Estimated ~15,000-25,000 companies globally fit this profile. At $1,000-3,000/mo, that's a $180M-900M TAM. Realistic SAM is probably $50-150M. Not a billion-dollar market on its own, but large enough for a very profitable SaaS business. Could expand into broader service health/governance over time.
$1,000-3,000/mo is reasonable for platform engineering budgets, but there's a challenge: the value is preventative (avoiding future incidents), not immediate ROI. Harder to sell than tools with direct productivity gains. However, after one major orphan-caused outage (which is inevitable), the buyer urgency becomes very high. The 'insurance' framing works but lengthens sales cycles. Cortex and OpsLevel prove this buyer exists and pays at similar price points.
Core MVP (git analysis + dashboard) is buildable in 4-8 weeks by a strong solo dev. Git APIs are well-documented. The hard parts: (1) HR/org data integration is messy — every company uses different HRIS (Workday, BambooHR, Rippling, etc.) and access is politically sensitive, (2) incident tool integrations (PagerDuty, OpsGenie, Datadog) add scope, (3) defining the 'orphan score' algorithm requires iteration and domain expertise. A credible MVP can ship with just GitHub/GitLab integration + manual org data upload, deferring HR API integrations.
This is the strongest dimension. Every existing platform (Cortex, OpsLevel, Backstage, Port, Datadog) treats ownership as a static metadata field. None compute ownership dynamically from git activity. None cross-reference with HR data to detect departed maintainers. None produce an 'orphan risk score.' CodeScene does git-based knowledge analysis but stops at the code level — no service-level operational risk. The gap is real, well-defined, and validated by the failure mode these platforms all share.
Natural SaaS subscription. Orphan detection is inherently continuous — people leave, services drift, dependencies go stale. The dashboard needs constant monitoring. Once integrated into org workflows (ownership assignment, risk reviews), switching costs are high. Could upsell into service health governance, compliance reporting, M&A due diligence for tech assets.
- +Clear, unaddressed gap — no competitor computes ownership dynamically or integrates HR data for orphan detection
- +Strong emotional resonance — every experienced engineer immediately understands this pain and has lived it
- +Natural wedge into the large and growing platform engineering / developer portal market
- +High switching costs once integrated into org workflows and ownership assignment processes
- +Preventative value proposition aligns with executive risk management concerns (SOC2, incident reduction, bus factor)
- +Can start narrow (orphan detection) and expand into broader service governance over time
- !Cortex, OpsLevel, or Port could ship an 'orphan detection' feature in a quarter — your differentiation is a feature, not a platform, until you expand scope
- !HR/org data access is politically sensitive and technically fragmented — the integration that makes this magical is also the hardest to get approved by customers
- !Preventative/insurance value proposition means longer enterprise sales cycles and harder ROI justification vs. productivity tools
- !Target buyer (platform engineering leader) is already overwhelmed with developer portal tooling pitches — cutting through noise is hard
- !Algorithm accuracy matters enormously — false positives (flagging healthy services as orphaned) will erode trust fast
Internal developer portal with service catalog, scorecards for service maturity, and operational readiness tracking. Scores services against compliance standards and surfaces ownership gaps.
Service ownership platform with maturity rubrics, auto-detection of services from K8s/cloud infrastructure, and self-service developer workflows.
Open-source developer portal framework with a software catalog at its core. CNCF incubation project with 100+ plugins. Teams register services via YAML with ownership metadata.
Behavioral code analysis tool that mines git history to identify code health hotspots, knowledge distribution, and 'knowledge islands' where only one person understands a codebase.
Service catalog built into Datadog's observability platform. Auto-discovers services from APM traces and enriches them with ownership, on-call, and runbook metadata. Includes scorecards for maturity.
GitHub/GitLab integration only. Connect to a single git provider, analyze commit history across repos, and compute an orphan risk score per repo/service based on: (1) last commit recency, (2) contributor concentration (bus factor), (3) contributor activity trend (declining = risk), (4) dependency staleness via lockfile analysis. Manual CSV upload for org data (who left, team mappings) instead of HR API integration. Simple dashboard showing ranked list of at-risk services with sparkline trends. One-click 'assign owner' workflow that creates a GitHub issue. Skip incident tool integrations for MVP — git signals alone are 80% of the value.
Free tier: connect 1 git org, scan up to 50 repos, basic orphan score → Paid ($1,000/mo): unlimited repos, org data upload, ownership workflows, Slack/Teams alerts, weekly digest emails → Enterprise ($3,000+/mo): HR system integrations (Workday, BambooHR), incident tool correlation (PagerDuty, OpsGenie), SSO/SAML, API access, custom scoring rules, compliance exports
8-12 weeks to first paying customer. Weeks 1-5: build MVP with GitHub integration + dashboard. Weeks 5-8: private beta with 3-5 platform engineering leaders from personal network or Reddit/HN community (the pain-signal audience). Weeks 8-12: iterate on scoring algorithm based on feedback, close first paid design partner at $500-1,000/mo. The Reddit thread itself is a lead source — commenters who shared war stories are potential early adopters.
- “the engineer who originally built it is long gone”
- “bandaid project for contractors in recent years”
- “no department actually wanted to maintain it, it was left to rot”
- “original context was lost”
- “a server sitting in a closet in Zanzibar”