What a modern meter-to-cash data platform looks like
Utility billing systems hold the truth about revenue and hide it well. The fix is not another extract — it is one model that finance, customer operations and the field all report from.
21 July 2026 | 7 min read | By Mela Consulting
If you have worked inside a utility's customer information and billing platform, you know its two personalities. It is scrupulously correct about every meter read, rate and adjustment, and it is almost impossible to report from. The tables are designed for transactions, not questions. So the organization grows a shadow layer of extracts, workbooks and departmental databases, each one a slightly different version of the truth.
The shape of the problem
Meter-to-cash crosses at least four systems — meter data, billing, payments and collections, and the general ledger — and at least three organizations: customer operations, finance and field services. Every hand-off between them is a place where consumption goes unbilled, bills are estimated for too long, or receivables age without anyone owning them. The cost is real and it is invisible on a monthly close.
What the platform needs to do
- Land billing, meter and payment data on a schedule the business can rely on, with change tracking rather than full reloads.
- Model service points, accounts, premises and meters once, with the relationships the billing system knows but never exposes.
- Define revenue, unbilled, exceptions and aging in one place, owned by finance and customer operations together.
- Surface exceptions daily — estimated reads beyond policy, service points with no active account, stalled bills — and route them to the team that resolves them.
- Serve executive, operational and regulatory reporting from the same model, so the numbers agree by construction.
Technically this is achievable with today's platforms: a lakehouse or warehouse on Microsoft Fabric, Azure or Databricks, incremental extraction from the billing database, a semantic model with named owners and Power BI on top. The hard part is organizational: agreeing the definitions and moving reporting off the extracts. That is where experienced help pays for itself.
The numbers agree by construction, not by reconciliation.
Once the model exists, automation follows naturally. Exception queues become workflows. Customer correspondence about estimated bills gets generated and routed. Regulatory reports become a scheduled output rather than a quarterly project. The platform stops being a reporting tool and becomes the operating layer for the revenue cycle.