Scott Engstrom is chief commercial officer at GridX, where he works with utilities on rate design, modeling and complex billing implementation.
Imagine if a utility could design a new dynamic rate on a Monday, assess its impact across every customer in the service territory the next day, and have it live on customer bills two months later. This is the pace at which utility billing needs to work because the rates, programs, and business models that will deliver affordability and demand flexibility can’t wait two to three years for the billing system to catch up.
Rate design used to be straightforward: collect enough revenue to cover costs and provide a return on capital. A once-in-a-generation surge in load growth, combined with continued investment in renewable generation, has changed what utilities and regulators ask of rates. They are increasingly a behavioral tool — a way to send price signals that shift consumption, reward flexibility, and protect affordability. The result is a rate portfolio that looks nothing like the one most billing systems were designed for: time-of-use pricing, critical peak pricing, demand charges, export compensation, EV-specific rates, and customer-specific riders.
According to a recent Guidehouse analysis commissioned by GridX, the number of approved electricity rates in the U.S. has grown roughly an order of magnitude over the past five years, to more than 50,000. Most customer information systems, or CIS, can’t natively express this rate complexity, so it gets pushed somewhere else: into meter data management system, or MDMS, configuration, custom CIS modifications, external calculation tools, or — more often than the industry likes to admit — spreadsheets maintained by a small group of people who understand how the tariff really works.
The volume is only half the story; the other half is how long each new rate takes to implement. Pilots make the problem especially visible: the whole point of a pilot is to get to market quickly, gather feedback and validate the design, but stakeholders wait so long for go-live that the learning loop breaks, and implementing a pilot rate costs roughly the same as implementing a rate for millions of customers.
Most utility billing systems were designed in the analog-meter era, when a bill was one price multiplied by one usage total. Those simplified, summarized numbers — billing determinants — are still generated in the MDMS today so the CIS can keep performing simple calculations on a small amount of data. That made sense when a residential customer had a fixed charge, a volumetric rate, and maybe a seasonal tier. The CIS didn’t need to understand the rate; it just needed to multiply set numbers. On the surface, it works.
Underneath, it creates a hidden tax on every rate change. Rate logic lives in both systems, so every new tariff is programmed twice, tested twice, and can fail in twice as many places. For a portfolio of time-varying, DER-aware, customer-specific rates that change every regulatory cycle, that compounds fast. The architecture that made billing reliable in the era of flat rates is the same one that makes it slow and costly to change in the era of complex ones.
The bottleneck shows up as delay. When a regulator approves a new rate with a fixed deadline, the constraint is how long it takes to implement the rate in the CIS, validate it across edge cases, and stand up the customer communications that go with it. Utilities have repeatedly testified that development timelines and testing capacity are among the largest practical constraints on rate innovation. Eighteen to 36 months to get a complex rate into production isn’t unusual but it’s a problem when commissions are ordering pilots on twelve-month timelines and demand flexibility is needed in quarters, not years.
The downstream effects compound. Virtual power plants don’t scale when settlement isn’t transparent enough for aggregators to commit capacity. Electric vehicle managed-charging rates don’t earn enrollment when customers can’t see what they’ll save. Community solar allocations don’t grow when the utility can’t support more subscribers without adding headcount. In each case, the program and the customer are ready but the billing layer isn’t.
Throwing tens of millions into the CIS misses the point
The reflexive response is to modernize the CIS, through custom enhancements or full-scale replacement. For any utility of meaningful scale, a replacement runs north of $100 million and often well beyond. Guidehouse documented one West Coast utility’s billing modernization request exceeding $700 million, with portions disallowed by regulators for insufficient benefit justification.
These aren’t outlier projects; they’re the pattern. Postmortems are remarkably consistent: complex rate scenarios account for many post-go-live billing defects, regardless of which CIS the utility relies on. Replacement doesn’t eliminate complexity risk. It often concentrates it at cutover. And embedding each new tariff into the CIS as custom code creates a maintenance liability that grows every regulatory cycle, hardest to justify for pilots, which is exactly where speed matters most.
The more useful question isn’t whether to overhaul the CIS. It’s which capabilities require a core system upgrade, and which don’t. Customer data, payments, receivables, and bill presentment are core CIS jobs that benefit from stability. Complex rate calculation is a different challenge with a different operating tempo. It needs configuration rather than custom code, and auditable outputs that can be defended to regulators before a rate goes live rather than after the first cycle of complaint calls.
There is more than one way to get there: modular rate engines that run alongside the CIS, rating modules modernized by CIS vendors themselves, or shared industry tooling. Whatever the mechanism, the principle is the same; decouple rate calculation from the systems that need to stay stable. New tariffs become a configuration exercise rather than a customization project, each one tested against real customer data before it touches production billing. And the same calculation that prices the bill can run the what-if analysis that supports rate design: utilities can model a proposed rate across every account in the territory and walk into the rate case with the impacts already in hand.
Each piece ties back to affordability. Faster deployment means demand-flexibility programs reach customers sooner, reducing the peak capacity utilities need to build. Lower implementation cost means less capital flowing into IT projects whose benefits are hard to defend in a rate case. Fewer billing defects mean fewer cancel-rebills and fewer disallowances. None of this is abstract; it shows up on the bill.
Why this matters now
Two trends make the status quo increasingly untenable. The first is performance-based regulation. At least 13 states are now tying allowed return on equity to performance metrics that include customer satisfaction, affordability, and time-varying rate enrollment. Billing accuracy and transparency have moved from back-office operational concerns to shareholder-value concerns. A utility that can't explain a complex bill is leaving basis points of allowed ROE on the table. And the stakes are rising fast.
The second is data center load. As large-load interconnections intensify system peaks and squeeze capacity margins, the complex rate designs needed to manage them — real-time pricing, dynamic demand charges, sophisticated curtailment products — are exactly what legacy CIS platforms struggle with most. The number of rate structures requiring interval-level precision is growing faster than most utility IT roadmaps anticipated, and the gap between operational need and system capability is widening.
What regulators can do
The fix is not the industry’s alone; regulators hold several of the levers.
Commissions can require implementation timelines and billing-readiness testimony as part of rate approvals, so the lag between approval and go-live is visible and on the record. They can reform pilot cost recovery so a five-thousand-customer pilot doesn’t carry the implementation cost of a full rollout — the single biggest reason pilot economics fail. States adopting performance-based regulation can add time-to-implement and billing accuracy to the metrics that shape allowed returns, giving utilities a direct incentive to clear the queue. And commissions can push for machine-readable, standardized tariff formats, so each newly approved rate doesn’t have to be manually re-translated into every system that touches it.
The hard infrastructure problem facing the utility industry isn’t limited to generation, transmission, or interconnection. It’s the last mile of value communication: the bill. Every dynamic rate, every DER program, every electrification incentive, every affordability initiative eventually must be priced, validated, and explained. If the billing layer can’t keep up, none of the rest scales. That’s not because the policy is wrong. It’s because customers won’t see it, regulators won’t trust it, and operations teams will quietly route around it.
Naming the bottleneck is the first step to clearing it. Utilities and regulators that treat billing modernization as a strategic priority and not just an IT backlog item will have a material advantage as rate complexity continues to accelerate.
