Home › Guides

How to run a solar load audit from your meter data

Before sizing a solar system — or a battery, or backup generator — you need to know what you're actually trying to offset: how much energy a site uses, when it uses the most, and how consistent that pattern is. That's a load audit. It sounds like a big undertaking, but if you already have interval meter data or a stack of utility bills, most of the work is mechanical: clean the data, compute a few summary numbers, and look at the shape of demand over a day and a year.

What data you need

The best input is interval data — readings every 15, 30, or 60 minutes from a smart meter, a power logger, or a building management system, exported as CSV or Excel. Two columns are enough: a timestamp and a demand or energy value. If you only have monthly utility bills, you can still get a rough total-energy picture, but you'll lose the hour-by-hour shape that actually drives solar sizing decisions — a site that uses most of its energy at night looks very different, from a solar perspective, than one that uses it all at midday, even if both have the same monthly total.

Both cumulative meter readings (a running total that only goes up) and instantaneous demand readings (kW at each timestamp) are workable — WattSketch asks which kind you have when you map your columns, and converts cumulative readings into interval energy automatically.

Step 1 — Upload

Drop your CSV or Excel file in. Parsing happens entirely in your browser — nothing is uploaded anywhere, which matters if your data includes anything you'd rather keep private, like an address tied to an account number. If you don't have a file handy yet, a real 14-day sample dataset is one click away, so you can see the whole flow before committing your own data.

Step 2 — Map your columns

Point the tool at your timestamp column and your demand/energy column, confirm the unit (kW, kWh, W, …), whether readings are cumulative or instantaneous, and how your dates are ordered (day/month/year vs. month/day/year — a classic source of silently wrong results if guessed incorrectly, so it's always shown for confirmation, not assumed).

Step 3 — Cleaning

Real meter exports are never perfectly clean. The cleaning step handles four things, all shown to you rather than fixed silently in the background:

• Duplicate timestamps — an identical repeated reading is merged; two different values at the same timestamp are flagged as a conflict for you to resolve.
• Short gaps — a missing reading or two is linearly interpolated, and marked as such everywhere it's used, so you can always tell a real measurement from a filled-in one.
• Long gaps — a gap too large to interpolate honestly (a logger that was offline for days) is left as a real hole in the data, not quietly papered over.
• Spikes — a statistical outlier relative to its local time window is flagged for your review, but its value is never changed automatically; you decide whether it's real or a sensor glitch.

Step 4 — Read the analysis

Once cleaned, the Analyze screen gives you the numbers that actually matter for sizing a system:

• Total energy (kWh) over the audited period, and average daily energy — the figure WattSketch's own solar size suggestion is built from.
• Peak demand (kW) — the single highest reading, relevant for any generator or inverter sizing decision.
• Load factor — average demand divided by peak demand. A low load factor (a site with a few sharp peaks and otherwise flat, low demand) behaves very differently from a high one (steady, consistent demand all day) even at the same total energy.
• Hour × day heatmap and duration curve — the shape of demand across the day and across the week, and what fraction of hours exceed any given demand level.

Step 5 (optional) — Extrapolate a typical year

If your audit only covers a couple of weeks or a season, WattSketch can extrapolate your real, cleaned data into a full synthetic year — useful for anything downstream that wants a whole-year view. See what a typical-year load profile is and why it matters for the full explanation.

Why this feeds into solar sizing

Your average daily energy use, together with your site's real solar yield per kilowatt at its actual location and orientation, is what WattSketch's suggested system size is built from — not a flat rule-of-thumb. The next step, sizing solar (and a battery, if your grid isn't fully reliable), is covered in sizing solar and battery for a site with an unreliable grid and a generator.