Just wrapped up an audit that reminded me why we can never take meter readings at face value. Avista customer in Spokane Valley, small retail chain with 8 locations. Bills looked normal, usage patterns seemed reasonable, but something felt off about the demand charges at three of the locations. Decided to request actual meter data from Avista rather than just relying on the bill summaries. Turns out two meters had been reading 10% high for 14 months due to calibration issues, and one meter was occasionally double-recording demand spikes during certain weather conditions. Found $23K in overcharges that would have been impossible to catch just looking at bills. The lesson: when something doesn't feel right, go straight to the meter data. Bills are summaries - meter data is the truth.
Don't assume meter readings are always correct
Greg, great catch! Avista up here has been pretty good about meter accuracy in my experience, but I've seen similar issues with other utilities. PSE had a batch of smart meters installed around 2019 that were reading about 3% high on certain load profiles. Took them almost a year to identify and fix the problem. I learned to always compare actual usage to historical patterns and industry benchmarks. If a restaurant suddenly starts using 15% more power with no operational changes, that's a red flag worth investigating.
TEP here in Arizona had a similar smart meter issue a few years back. What really opened my eyes was when I started requesting 15-minute interval data instead of just monthly summaries. You can spot meter malfunctions, billing errors, and even equipment problems that don't show up in the monthly bills. Had one client where the meter was recording phantom demand spikes every Tuesday at 3 AM - turned out to be a faulty meter communication module that was adding its own power draw to the readings.
Alabama Power down here in Montgomery has been pushing smart meters hard, but the older mechanical meters are still pretty common in rural areas. Had a manufacturing client where the meter disc was sticking intermittently - would run slow for days then suddenly catch up all at once. Made the billing look erratic as hell. The utility actually credited back 18 months of overcharges once we proved the meter was defective. Sometimes the old-school visual inspection still catches things the fancy diagnostics miss.
National Grid here in Rhode Island has been pretty good about meter accuracy, but I learned never to assume. Had a small office building where the meter was installed backwards - literally wired in reverse. The usage readings were correct but the power factor measurements were completely wrong, causing massive reactive power penalties. Took six months to figure out because the kWh usage looked normal. Now I always verify meter installation as part of my site surveys.
OG&E here in Oklahoma City has had some issues with meter communication errors that cause estimated readings for months at a time. The bills show "estimated" but clients often don't notice or understand what that means. Had one case where they estimated high for 8 months, then did a massive adjustment that made it look like the client had a huge usage spike when it was really just a correction. Always look for estimated reading flags - they're opportunities for billing errors.
SWEPCO over here in Louisiana has some really old meters in industrial areas that can develop issues over time. What I've learned is to compare the meter's kWh readings with the client's actual equipment runtime and load calculations. If a 500 HP motor is supposed to draw about 375 kW at full load and runs 10 hours a day, but the meter shows wildly different usage, that's worth investigating. Basic sanity checks can catch a lot of meter problems.
MLGW here in Memphis keeps pretty good records, but I've seen meter accuracy issues crop up especially after storm damage or construction work near the meter. Had a client where a contractor accidentally bumped the meter housing and threw off the calibration by about 7%. The utility didn't catch it for 11 months. Now I always recommend clients take photos of their meter readings monthly, especially if there's any construction or maintenance happening nearby.
Duke Energy here in Charlotte has been upgrading meters like crazy, but the installation process sometimes introduces errors. Had two cases where the meter multiplier was set wrong during installation - one was set to 1x instead of 10x, the other was 100x instead of 10x. Both caused massive billing errors that took months to sort out. Always verify the meter multiplier matches the CT ratio and the tariff setup.
Ohio Edison up in Youngstown had a batch of meters that were losing time synchronization, causing demand readings to be recorded in the wrong time blocks. A client was getting charged peak demand rates for usage that actually happened during off-peak hours. The 15-minute interval data showed the problem clear as day, but you had to know to look for it. Saved the client about $15K annually once we got it fixed.
Excellent thread Greg. This is exactly the kind of detailed detective work that separates professional auditors from bill reviewers. I always tell new auditors that the meter is your primary witness - it sees everything that happens electrically at the site. But like any witness, meters can be wrong, corrupted, or tampered with. The key is learning to spot when the story doesn't add up and having the tools and knowledge to dig deeper. Great examples everyone!