How to Calculate Average Daily Electricity Usage from Meter Readings
Convert two dated cumulative readings into a fair daily rate, handle partial days, and avoid misleading calendar-month comparisons.
The short answer
Subtract the earlier cumulative reading from the later one, then divide by the actual elapsed time in days:
Average daily usage = (later reading − earlier reading) ÷ elapsed days
This gives a useful rate for the observed interval. It does not guarantee tomorrow's usage or explain which appliance caused it.
A complete worked example
- September 2 at 8:00 pm: 31,480 kWh
- September 9 at 8:00 am: 31,558 kWh
- Register difference: 78 kWh
- Elapsed time: 6.5 days
78 ÷ 6.5 = 12 kWh per day
Rounding the interval to “one week” would produce 11.14 kWh/day and understate the observed rate. For rough household monitoring that difference may be acceptable, but investigations and comparisons benefit from real timestamps.
Compare equal ideas, not just equal labels
“This week” can mean seven complete days, five days so far, or two meter readings taken eight days apart. “This month” may not match the provider's billing cycle. Always carry the start date, end date and covered hours with the rate.
| Comparison | Fair? | Why |
|---|---|---|
| 14 kWh/day over 7 days vs 12 kWh/day over 7 days | Usually | Same duration, if conditions are similar |
| 90 kWh in 5 days vs 120 kWh in 10 days | Not as totals | Normalize both to kWh/day first |
| September calendar month vs Aug 12–Sep 10 bill | Not directly | Different date windows |
| One hot weekend vs a mild monthly average | Weak | Weather and occupancy differ |
Handling readings that cross your chosen boundary
Suppose you want September 1–30 but your nearest readings are August 29 and October 2. You can calculate the exact usage between the observations. Allocating part of that usage into September requires an estimate unless finer readings exist.
Say so plainly: “about 11 kWh/day based on observations spanning Aug 29–Oct 2.” Do not present an interpolated calendar total as a physical reading.
When a daily average is useful
- Checking whether cycle pace is rising before the bill arrives.
- Comparing intervals of different lengths.
- Establishing a normal range for the same property and season.
- Deciding whether a more controlled appliance test is worthwhile.
- Spotting a data-entry error that implies impossible overnight use.
When it is misleading
A daily average hides timing. A home using 70 kWh over seven days averages 10 kWh/day whether the usage was steady or one 40 kWh event caused most of it. It also hides which meter contributed when several meters are combined.
Use the average as a summary, then inspect the underlying dated intervals. Meter contribution and shorter reading intervals provide the context an average cannot.
A repeatable MeterPilot workflow
Choose the property and meter, record cumulative readings with their actual timestamps, and use a report range that states its dates. If an interval only partly overlaps the requested range, keep “about” or estimated coverage visible. Compare against the same meter rather than a portfolio total unless the portfolio is the decision you are making.
Reserved for a report state that shows cumulative boundaries, elapsed time and the resulting daily rate together. No unrelated screenshot has been substituted.
Compare a defined interval—not a vague “normal”.
Build a dated history in MeterPilot and compare like-for-like periods with coverage and meter context visible.
See the MeterPilot workflow