Try a simple exercise: count the places where your organisation's communication identifiers are recorded. Telephone numbers in carrier portals — usually several, one per carrier, per country. SIP identities in UC admin consoles. Messaging sender IDs in marketing and developer platforms. SIM and eSIM profiles in device management tools. And, holding it all together, spreadsheets — maintained by whoever last needed an answer.
Each of these records is roughly accurate about its own fragment. None of them is authoritative about the estate. And the gaps between them are where the cost accumulates.
What fragmentation actually costs
Finance pays first and most visibly. Identifiers that are billed by a carrier but no longer used by anyone. Identifiers allocated to a customer but never invoiced. Ranges reserved years ago for projects that finished, or never started. A first governed reconciliation routinely surfaces a material share of the estate in this state — cost that no single inventory could see, because each system was only wrong about the part it couldn't check.
Operations pays continuously, in smaller increments. When no record is authoritative, every change begins with research: which carrier holds this number, who owns it internally, what services will break if it moves. Multiply that by every port, every provisioning request, every routing change. The cost becomes unmissable at migration and M&A moments, when "discover what we actually have" becomes a project phase of its own — and often the longest one.
Compliance pays under the worst conditions: on a deadline, facing backwards. When an auditor or regulator asks who authorised a change, or a fraud investigation asks who controlled an identifier on a given date, the evidence has to be reconstructed from systems that were never designed to produce it. The estate's history exists only as fragments of fragments.
Why the obvious fix fails
The instinctive response is a consolidation project: pull every inventory into one master spreadsheet or database, reconcile the conflicts, declare a single source of truth. It works — for about a week.
The problem is that a consolidated inventory is a photograph, and the estate keeps moving. Numbers port. Services migrate. New sender IDs get registered by a team that never heard about the master list. Nothing forces changes through the consolidated record, so the record starts decaying the day it's finished. Visibility without governance isn't a source of truth; it's a snapshot of one, ageing.
What a governed estate changes
The durable version inverts the relationship between the record and the change. Instead of the record chasing what happened, changes go through the record: an identifier can't move, activate or retire without its ownership, state and policy being checked — and updated — as part of the change itself.
That requires a different foundation than a spreadsheet. The Identifier Graph — the governed model of every communication identifier and the relationships between them, providing a single source of truth for ownership, policy, lifecycle and execution — is that foundation: not another inventory alongside the others, but the layer the others connect to.
The per-function economics then invert too. Finance reconciles once, and the reconciliation stays true. Operations queries instead of researching — the answer to "who owns this, what depends on it" is a lookup, not an investigation. And compliance stops reconstructing evidence, because evidence is produced as a by-product of every governed change, not assembled after the fact.
What to read next
This is one symptom of a broader gap — for the full argument that identifier governance is its own discipline, start with what Communication Identifier Governance actually means. For the operator's revenue-focused version of the same case, read why governance, not numbering, is the constraint on growth. And for how the platform is built to hold the governed estate, see the architecture overview.