What a Utility Network Migration Actually Costs (And What's Actually Yours to Control)
The honest breakdown of what an ArcGIS Utility Network migration really costs a utility your size — what's genuine spend, what you already own, and why most quotes price a solved problem as if it were a mystery.
Search "ArcGIS Utility Network migration cost" and you'll find the same answer everywhere: somewhere between $500K and $1.5M, with a pile of line items that never quite add up to anything you can act on. Licensing here, implementation there, data migration somewhere in the middle. Big numbers, no floor, no ceiling, no sense of what you're actually paying for versus what you already own.
That's not an answer. That's a shrug with a dollar sign on it.
If you run GIS for a utility in the 10,000–150,000 meter range, here's the more useful version: what you're actually going to spend money on, what you're not, and which parts of this project are genuinely yours to control.
The number that shouldn't be in this conversation at all: licensing
Almost every cost breakdown you'll find leads with licensing — often $150K–$400K a year — presented like it's a new bill arriving because of your UN migration.
For the overwhelming majority of utilities already running Esri, it isn't. If you have an Enterprise Agreement, you're already paying for ArcGIS across dozens of departments — planning, engineering, operations, maybe even your public-facing outage map. The Utility Network doesn't sit outside that agreement charging you extra; it's part of what you've already licensed. Migrating to it doesn't create a new subscription. It changes how you use the one you have.
So when a vendor's cost estimate opens with a six-figure annual licensing number, ask what it's actually measuring. If you're already an Esri shop, that number belongs in a totally different conversation — your enterprise agreement renewal — not in the cost of this project. Leading with it just makes the whole migration look scarier than it is.
What you're actually paying for
Strip licensing out, and a UN migration for a utility your size comes down to four real cost centers:
1. Infrastructure. Servers, environments, and the underlying platform setup — whether you're standing this up on-prem or in the cloud. This is real, necessary work. It's also the one piece of this list that's largely outside a migration partner's control — it's yours, or your IT department's, or a hosting decision you make independently of who does your data work.
2. Data modeling. Deciding how your network actually gets represented in the new schema — asset types, domains, subtypes, association rules. Get this wrong and everything built on top of it inherits the mistake.
3. Data migration. The actual work of moving your network — conductors, devices, structures, connectivity — out of your legacy geometric network or ArcFM environment and into the new model.
4. Data cleanup. The unglamorous part nobody puts a number on upfront: fixing the decades of small inconsistencies, orphaned features, and broken connectivity that every utility's GIS accumulates — because the new model won't tolerate what the old one quietly ignored.
Infrastructure aside, these three — modeling, migration, cleanup — are where the real spend lives. And they're also where you, as the utility, have the most leverage over the outcome.
Why the real dollars — and the real risk — live in one place
Here's what the industry-wide cost ranges get wrong for a utility your size: they're built off enterprise IOU projects with sprawling territories, heavy integrations, and years-long timelines. A lot of that cost is inherent to scale you don't have.
But the modeling–migration–cleanup work isn't proportional to customer count the way people assume — and it's not the mystery it's usually priced as, either. The data quality problems every utility's GIS accumulates — broken connectivity where old workarounds papered over topology, orphaned features, inconsistent schema, decades of small inconsistencies nobody had time to fix — aren't unique to your system. They're remarkably consistent across the industry. A 20,000-meter utility and a 200,000-meter utility tend to have the same handful of issues; the larger utility just has more of them to fix.
That repeatability matters, because a traditional systems integrator still prices this work as if every engagement starts from zero — diagnosing your specific mess for the first time, billing by the hour to figure out what's wrong before they can even start fixing it. That's where a lot of the six-figure inflation actually comes from: not the work itself, but re-solving a problem that's already been solved a hundred times over, on someone else's clock.
That's the frustrating part of the conversation nobody has honestly: the price isn't scaled to you, and the uncertainty in most quotes isn't really about your data — it's about a staffing model that treats known problems like unknowns every time.
The part we've actually solved
This is where we've spent our effort at Nutility: not on infrastructure, and not on reinventing data modeling practices from scratch for every client, but on the pattern recognition itself. We've trained our AI agent workforce on the common data quality issues that show up across geometric network migrations — the specific connectivity breaks, the specific schema mismatches, the specific remediation actions each one calls for — so the diagnosis isn't a billable mystery each time. It's a known playbook, applied at whatever volume your network requires.
That's also why our cost calculator scales with meter count the way it does. It's not because a bigger network is riskier or less predictable — it's because there's simply more of the same, known work to do.
Effort scales with volume. It doesn't scale with uncertainty.
So what does it actually cost, for a utility your size?
That's a fair question, and it deserves a real number — not another paragraph of hedging.
It depends mainly on your domain (electric, gas, or a multi-domain system) and your network size — the two things that actually drive volume — which is exactly why we built a calculator instead of publishing one more misleading blanket range. Eight quick inputs, no login, and you get a budgetary number you can actually take to your board.
Estimate your migration cost →
If you want a number precise enough to budget against, not just a range, that's what the readiness assessment is for: we read your actual geometric network geodatabase and score it — schema, connectivity, data quality — and hand back a fixed, not-to-exceed price before anything else starts. It's a flat $10K, fully credited toward your migration if you move forward.
And that price holds. Any data quality issue we find during that assessment that would inhibit our commitment to provide you a connected and traceable network in the new model — we fix it, as part of the scope we already quoted. Not a change order six weeks in. Not a "we found more problems than expected" conversation. If it's standing between you and a connected, traceable network, it's already priced in.
Wherever you're at in planning your migration to the Utility Network, we're here to help — whether that's creating a migration estimate, evaluating your source data with the most substantial readiness assessment in the industry, or getting started on your migration outright.
All of it happens in one place. Create a login at the top of the site and you're in your own dashboard — the same one you'll use to initiate work orders and track status once your project's underway, not a generic portal shared across every client. Run your numbers, review your readiness assessment, sign, and start your project, all without leaving the platform.