- Esri Geometric Network retires with ArcMap in March 2026 - no extensions, no exceptions
- Realistic migration timeline: 6-18 months (not the 3-6 months in Esri's marketing materials)
- Cost per asset class: $50K-$200K depending on network complexity and data quality
- 3 viable options: full migration, minimal migration (keep running but unsupported), or open-source alternative
Geometric Network retires in March 2026. Thousands of utilities worldwide are facing a forced migration to Utility Network. Esri's guidance says 3-6 months. After working on network migrations for over a decade, the honest answer is 6-18 months. Here is what they are not telling you.
The Deadline: What Is Actually Happening
ArcMap retires in March 2026. This is not a soft deprecation or a gentle nudge. Geometric Network only exists in ArcMap - there is no equivalent in ArcGIS Pro. When ArcMap goes, Geometric Network goes with it.
WHAT THIS MEANS IN PRACTICE
- -Geometric Network data does not disappear, but it becomes read-only
- -No edits, no traces, no new analysis on legacy networks
- -No security patches, no bug fixes, no technical support
- -Esri's position: migrate to Utility Network or lose network analysis capability
The real concern is not that your data vanishes overnight. It is that you are running critical infrastructure management on software that nobody is maintaining. Every month you delay increases your technical debt and your risk.
What Actually Changes: Geometric Network vs Utility Network
Utility Network is not just a renamed Geometric Network. The architecture is fundamentally different. Understanding these differences is critical to scoping your migration accurately.
| Capability | Geometric Network | Utility Network |
|---|---|---|
| Platform | ArcMap only | ArcGIS Pro + Enterprise |
| Network topology | Simple connectivity | Structure networks + domains |
| Subnetwork management | No | Yes (circuits, pressure zones) |
| Dirty areas tracking | No | Yes (incremental validation) |
| Asset group/type | Feature class based | Attribute-driven |
| Network rules | Connectivity only | Connectivity + attachment + containment |
| Tracing | Basic upstream/downstream | 13+ trace types |
THE HONEST TAKE
Utility Network is genuinely more capable. The question is not whether it is better - it is. The question is whether the migration cost and timeline are worth it for your network, given your specific complexity, budget, and strategic direction.
Real Migration Timeline (Not Esri's Version)
Esri's marketing materials suggest 3-6 months for a Utility Network migration. That timeframe is possible for a small, clean network with no custom tools and no integration dependencies. For a real utility with 500K+ assets, 15+ years of data, and 20+ custom ArcMap tools? Budget 6-18 months.
REALISTIC MIGRATION PHASES
Assessment and Data Audit
4-8 weeksInventory all Geometric Network feature classes, connectivity rules, custom tools, and third-party integrations. This is where most teams underestimate - data quality issues surface here.
Data Model Design
4-6 weeksMap Geometric Network features to Utility Network asset groups and types. Define domain networks, tiers, subnetwork definitions. This requires deep domain knowledge - not just GIS skills.
Data Migration
6-12 weeksThe hardest phase. Feature migration, connectivity rebuilding, rule configuration. Expect 2-3 iterations of migrate-test-fix cycles. Data quality issues from the Geometric Network will surface as hard failures.
Custom Tool Migration
4-8 weeksEvery ArcMap toolbar, every custom ArcPy script, every ModelBuilder model that touches the network must be rewritten for ArcGIS Pro SDK or ArcPy 3.x. This is often 40% of total effort.
Testing and Validation
4-6 weeksRun every trace, every analysis, every report against the new network. Compare results to Geometric Network outputs. Expect discrepancies - Utility Network traces differently.
Training and Rollout
2-4 weeksEvery field crew, every analyst, every manager needs training. Utility Network workflows are fundamentally different. Budget for a productivity dip.
TOTAL REALISTIC TIMELINE
Complex networks with extensive custom tooling can push to 18 months
If you are considering migrating custom ArcPy tools to open source as part of this transition, our ArcPy to GeoPandas translation guide covers the practical steps.
Cost Per Asset Class
Migration costs vary dramatically by asset class. The complexity of your network model, the quality of your data, and the number of custom tools all drive cost. For a detailed breakdown of Utility Network licensing costs specifically, see our complete Utility Network cost guide.
MIGRATION COST BY ASSET CLASS
Includes: data migration, model design, custom tool migration, testing
Excludes: Esri licensing, hardware upgrades for ArcGIS Pro, training time
THE HIDDEN COST: CUSTOM TOOLS
A utility with 30 custom ArcMap toolbars faces 4-8 weeks of pure development work at $150-$250/hr for a qualified ArcGIS Pro developer. That is $50K-$100K just for tool migration - before you touch a single data record.
5 Common Migration Failures
These are the failure patterns we see repeatedly. Every one of them is avoidable with proper planning.
Data quality assumptions
"My Geometric Network is clean." It is not. Fifteen years of edits without topology validation means orphaned features, disconnected networks, and invalid geometries. Budget 20% of migration time for data cleanup alone.
Underestimating custom tools
Every Python script, every ModelBuilder model, every third-party extension that touches the Geometric Network must be rewritten. Esri provides no automated migration path for custom code. This is routinely 40% of total project effort.
Trace result differences
Utility Network traces use different algorithms. Results will differ from Geometric Network in edge cases. Field crews who have relied on traces for 10+ years will notice, and they will escalate. Plan for a reconciliation period.
Enterprise architecture gaps
Utility Network requires ArcGIS Enterprise with specific portal configuration. Many utilities running ArcGIS Server + Geometric Network do not have the required Enterprise stack. This is a separate infrastructure project on top of the data migration.
Change management failure
Analysts who have used Geometric Network for a decade resist the workflow changes. Training is not optional - it is critical path. Expect a 15-30% productivity dip for the first 3-6 months after go-live.
Alternatives to Full Migration
Full Utility Network migration is not the only option. Depending on your situation, one of these alternatives may be more appropriate.
Full Utility Network Migration
$100K - $500K+
6-18 months
Utilities needing subnetwork management, advanced tracing, long-term Esri commitment
Minimal Migration (Stay on Geometric Network)
Keep ArcMap running on isolated machines for network operations. No security patches, no support, no new features. Increasing compatibility issues over time.
$0 upfront, increasing technical debt
High and growing
Small utilities with simple networks and low budget
Open-Source Alternative (PostGIS + pgRouting)
Migrate network data to PostGIS, use pgRouting for network analysis. You lose Esri-specific features (dirty areas, subnetworks) but gain platform independence and eliminate licensing costs.
$30K - $100K (one-time) + ongoing maintenance
Lose Esri ecosystem, gain independence
Utilities questioning long-term Esri commitment
Hybrid (Automate Workflows, Migrate Later)
Automate the most time-consuming workflows first using cloud pipelines, defer full network migration. Delivers immediate efficiency gains while buying time for a considered platform decision.
$50K - $150K
2-4 months for initial automation
Utilities needing immediate gains but not ready for full migration
When NOT to Migrate Yet
Not every utility should rush into this. Here are five scenarios where delaying or choosing an alternative path is the smarter decision.
You are within 6 months of the deadline with no plan
Starting now means rushing. A rushed migration creates more problems than running unsupported for 6-12 months while doing it properly. Get the plan right first.
Your Geometric Network is your only source of truth
If the migration fails, you need rollback capability. Do not migrate without a tested backup and rollback plan. Verify your backups actually restore before starting.
Your team is already overloaded
Migration requires dedicated resources. Doing it "on the side" of regular work doubles the timeline and triples the risk of errors. Either free up people or wait until you can.
You are evaluating Esri alternatives
If there is any chance you are leaving Esri in the next 3-5 years, investing $200K+ in Utility Network is questionable. Consider the open-source path or the hybrid approach instead.
Your network is simple
If you have fewer than 50K assets and no subnetwork requirements, PostGIS + pgRouting is genuinely simpler and cheaper. Do not spend $200K on Utility Network when a $30K open-source solution covers your needs.
Decision Framework
Use these three questions to determine your path forward.
QUESTION 1
Do you need subnetwork management (switching, isolation, pressure zones)?
Utility Network is likely the right choice. Its subnetwork model is far ahead of any alternative.
Consider PostGIS + pgRouting. Basic network tracing does not require Esri's full stack.
QUESTION 2
Is your Esri spend greater than $200K/year?
Migration ROI is strong. You are already invested in the platform. Payback within 2-3 years.
Run the cost comparison carefully. At lower spend levels, open-source may deliver better value.
QUESTION 3
Do you have 6+ months before you need the migration complete?
Start planning now. Assessment phase alone takes 4-8 weeks, and early planning reduces cost.
Consider Option B (minimal migration) as a bridge. A rushed migration is worse than a planned delay.
Frequently Asked Questions
When does ArcGIS Geometric Network retire?
Geometric Network retires with ArcMap in March 2026. After this date, it will no longer receive updates, security patches, or technical support from Esri.
How long does Utility Network migration take?
Realistically, 6-18 months for a mid-sized utility with 500K+ assets. This includes assessment (4-8 weeks), data model design (4-6 weeks), data migration (6-12 weeks), custom tool migration (4-8 weeks), testing (4-6 weeks), and training (2-4 weeks).
How much does Utility Network migration cost?
Migration costs vary by asset class: water distribution ($50K-$100K), electric distribution ($100K-$200K), telecom ($150K-$300K). These costs cover data migration, model design, and custom tool migration but exclude Esri licensing and hardware.
Do I have to migrate to Utility Network?
No. You have alternatives: continue running ArcMap unsupported (increasing risk), migrate to open-source (PostGIS + pgRouting), or take a hybrid approach (automate workflows first, migrate later). The right choice depends on your network complexity and long-term platform strategy.
Get Workflow Automation Insights
Monthly tips on automating GIS workflows, open-source tools, and lessons from enterprise deployments. No spam.