AI That Closes the Gap Between Field and Finance: PowerPlan Intelligent In-Service

AI That Closes the Gap Between Field and Finance: 

PowerPlan Intelligent In-Service

Every month, finance and accounting teams at asset-intensive companies face the same scramble. Work has been completed in the field. Capital has been deployed. Assets are economically active. But somewhere between the job site and the general ledger the data didn’t make it, and now teams are chasing project managers, manually reviewing hundreds of work orders, and letting the unitization backlog grow so that close timelines don’t slip. 

When we polled utility accounting professionals in a webinar on this topic, 93% told us that work order data gaps impact their close to some extent, with more than a quarter dealing with it every close cycle. Another fifth said gaps were rare, but when they did occur, the consequences were painful. These consequences include delayed rate base recognition, distorted depreciation, audit exposure, and manual rework that eats into every close cycle. 

How Work Order Data Gaps Create Financial Risk for Utilities 

The downstream impact of a single data gap is easy to underestimate. An asset placed into service but lacking a recorded completion or in-service date not only creates a reconciliation headache, but it can cascade into asset valuation errors, depreciation distortions, incorrect tax basis, rate base inaccuracies, and ultimately audit findings that surface long after the original work was done. 

In the same poll, 69% of accounting professionals said their biggest work order data pain wasn’t one specific issue—it was all of them, varying by cycle. These issues include incomplete project estimates, missing or late in-service and completion dates, and unrecorded retirements. The combination compounds the risk and the workload. 

Finance teams carry the burden every close cycle. Tax teams deal with incorrect basis and missed retirements. Regulatory teams file using data that may not accurately reflect rate base.  

Find Data Gaps, Fix Them, and Prevent Them from Happening Again 

Embedded in the PowerPlan NXT platform, Intelligent In-Service is our answer to the data gaps that create undue burden on finance and accounting teams and the threats those gaps pose from a financial and regulatory perspective. One of the components of Intelligent In-Service is the Project Date Predictor feature, available today: AI-predicted in-service and completion dates for work orders with missing data. 

Four key capabilities of Project Date Predictor guide users from detection through resolution of work order gaps:  

• AI agents run automatically on a monthly cadence or on demand, scanning work order and asset records to surface data gaps that create close cycle risk. Instead of discovering problems during reconciliation, your team sees them before close begins. 
• Flagged data gaps are prioritized by level of impact: financial significance, time-sensitivity, and AI confidence level. Your team works the list in priority order, not discovery order—focusing on the most pressing gaps first. 
• Every flagged record comes with full context so you can drill into the details behind every prediction, understand what drove it, and accept or reject the findings.  
• Trend and pattern reporting surfaces upstream behaviors that create gaps in the first place, giving operations and accounting a shared view of where the process breaks down. 

Every flagged record comes with full context. You can drill into the details behind any prediction, understand what drove it, and make the call to accept the recommendation, adjust it, or set it aside. Nothing changes in the system until a human decides it should. 

Reduce Close Cycle Risk and Manual Rework

Before Intelligent In-Service’s Project Date Predictor functionality, missing in-service and completion dates were discovered during reconciliation, or worse, during audit. Now, AI-predictions are available to review and act on before close begins. As a result, teams spent hours pulling data together and chasing project managers for answers. Now, context and recommended action are delivered in one place, instantly. 

That moves your team from scrambling at quarter close to staying ahead all quarter. It means fewer late nights chasing data, more confidence in the numbers before the deadline, and less exposure when auditors come looking. 

For teams managing Construction Work in Progress (CWIP) and even aging Construction Completed Not Classified (CCNC) balances, the business case is direct: faster identification of assets eligible for capitalization accelerates rate base recognition. Administratively delayed but economically active assets get off the CWIP/CCNC backlog and onto the books where they belong. 

Human-in-the-Loop AI in Utility Asset Management 

Amidst the AI noise in the market right now, we took a simple, thoughtful, and forward-looking approach with Intelligent In-Service and our AI philosophy across the platform: AI should surface what your team needs to act on, with the context to act confidently—and it should never make decisions autonomously on your behalf.

That principle shaped every design decision. After Project Date Predictor analyzes historical work order activity patterns to identify records likely eligible for transition to “In Service” or “Complete” status, it surfaces those findings ranked by urgency, confidence level, and financial impact. Your team reviews the recommendations, drills into the details, and makes the call—accept, adjust, or defer. No data is modified without a human decision. 

Intelligent In-Service Built Alongside Customers 

We co-designed Intelligent In-Service with nine customer organizations across more than 50 hours of collaborative sessions in which we placed imperfect prototypes in front of real users, watched them work, and course corrected when the feedback told us where we weren’t headed in the right direction. 

The key lesson from that process was that data quality feeds AI quality. Starting narrow—focusing first on the most common, highest-impact upstream data gap before expanding—let us prove value quickly and build trust in the recommendations before scaling to more complex use cases. 

AI-Powered Utility Asset Management on the PowerPlan NXT Platform 

Intelligent In-Service is live on the NXT platform today with the Project Date Predictor capability. Customers opt in through feature flags and can adopt at their own pace.  

The friction that defines your current close process—the manual reviews, the data chasing, the reactive scramble—is solvable. Intelligent In-Service is one significant step toward solving it.

Want to see Intelligent In-Service in action? Request a personalized demonstration to explore how it can help your team close faster, with greater confidence, and less manual effort.