How we work out the cost

How a plan's cost is worked out here, in enough detail to check us. Everything below happens in your browser, against the plan data retailers publish.

Where the plan data comes from

The Consumer Data Right energy feed — the same product reference data behind the AER's Energy Made Easy. We are independent of the AER, Energy Made Easy and every retailer; we simply read the published feed. Plans are refreshed as retailers change them, and superseded versions of the same plan are collapsed so you see the current one rather than five near-identical rows.

Your usage: the part most comparisons skip

Most comparison sites price against a model household — a single yearly figure, or a handful of stereotypes. Two homes using the same total per year can owe very different amounts, because when you use it decides which rates apply.

So the site prices against a usage shape: 24 hourly values, optionally different for each day of the week. You can build that shape three ways.

• Import your meter data. A NEM12 file (and several retailer and inverter formats) is read in the browser. We take up to the last 12 months and build an average shape per day of the week, so seasonal and weekday/weekend patterns survive. Up to 18 months of readings are retained for the backtest below.
• Shape it by hand with the usage bars, starting from a low, medium or high preset.
• Enter a daily total and let the default shape spread it.

Solar is a separate export profile, because what you export matters at different hours from what you consume. Controlled load is separate again — it is a distinct circuit on its own tariff, not part of your general usage.

What gets priced

Each plan's published tariff is turned into a rate for every half-hour of a week, then your usage is priced against it across a full year. That includes:

• daily supply charges;
• single-rate, time-of-use and stepped/block tariffs, including seasonal variations and the season that applies today;
• free or discounted windows, which some plans use heavily;
• controlled load on its own rate;
• demand charges — measured over the plan's own measurement period, on the days and windows it specifies, in kW or kVA as the plan states;
• solar feed-in, including tiered and capped bonus rates, and plans that charge you to export rather than paying you;
• conditional discounts you tell us you qualify for (pay on time, direct debit);
• card payment surcharges, for the payment method you select;
• GreenPower, if you turn it on.

Prices are shown including GST. Consumption charges are grossed up because the published rates exclude GST; feed-in credits are not, because residential exporters are not paid GST on them.

The backtest: what you would actually have paid

If you import meter data, the site does something a model household cannot: it prices your real half-hourly readings, month by month, against each plan. That is a different and stronger number than an annual estimate, because it bills your actual peaks rather than an averaged shape — and demand charges in particular are decided by peaks.

Where we round, and which way

Published plan data is sometimes ambiguous or incomplete. Two rules apply.

We estimate on the high side, never in the plan's favour. Where a charge could reasonably be read two ways, we take the more expensive reading. An estimate that flatters a plan is the one that costs you money.

We flag it rather than guess silently. Where an assumption was needed, the plan is marked as approximate and the assumption is spelled out in its details. If a figure cannot be derived at all, you get a warning rather than a confident wrong number.

What we cannot price

• Offers retailers do not publish — some retention deals, and some EV-specific plans available only on request.
• Conditional and one-off fees that depend on your circumstances. They are listed in the plan's terms rather than folded into the headline figure.
• Anything for Western Australia or the Northern Territory: the data set holds no plans there.

How plans are ordered, and what our commercial arrangements do and do not affect, is on our editorial policy. Common questions about meter files and imports are in the FAQ.