A commercial battery business case is one of the most asked and answered questions in WA energy right now. Asked by site operators looking at rising bills. Answered by the same brochures that have answered it for a decade. The answer in 2026 is not the answer in 2020.

We model commercial battery business cases as part of every site review. This piece is a generalised view of when a battery on a WA commercial site pays back, when it doesn't, and what changes the answer.

It's not an investment recommendation for any specific site. The variables that matter are site-specific. Read it as the framework.

What a commercial battery can earn

A commercial battery on a WA site can capture value across four revenue lines. The case for the asset is the sum of these, less the cost of the battery, the install, and the ongoing operation.

Demand charge reduction. This is the largest line on most WA commercial bills, and it's the line a battery moves most reliably. Western Power network charges include a demand component calculated against the site's peak in defined windows. A battery that discharges before the peak sets reduces that line by an amount equivalent to the kW reduction times the demand rate times twelve months. On a site with a $25 per kVA per month demand rate and a 100 kW peak reduction, that's $30,000 a year. On a site with a heavier demand profile, materially more.

Tariff arbitrage. WA time-of-use commercial tariffs reward charging the battery in off-peak windows and discharging into peak windows. The spread between off-peak and peak energy rates is widening as time-of-use tariffs sharpen. A battery cycling daily across this spread earns the cycle count multiplied by the spread multiplied by the throughput. For a 200 kWh battery cycling once a day at a 15c per kWh spread, that's around $11,000 a year, before round-trip losses.

Market participation. Eligible commercial batteries in the WEM can participate in the wholesale market and earn for system services. The participation depends on site size, metering, and the regulatory pathway being in place. For sites where the pathway opens, this is incremental revenue on top of the on-site value, not instead of it. Most commercial sites don't access this line today. The ones that will, will pull ahead.

Solar coordination. For sites with both solar and a battery, the battery captures solar that would otherwise be exported at low rates and uses it later against tariffs or peaks. The value depends on the solar size, the load profile, and the tariff position. On a site where solar generates more than it consumes in the middle of the day, this is meaningful. On a site where solar lines up with daytime load, less so.

The total annual return is the sum of these four lines, less the cost of operating the asset.

Three sites, three answers

Below is what the maths typically looks like across three site types we encounter regularly in WA. The numbers are illustrative, the variables are the point.

Site one: light commercial

A light commercial site running 50 kW to 100 kW peak. Operating hours 7 am to 6 pm. Existing solar of 50 kW to 80 kW. Mid-tier time-of-use tariff. Modest demand charges.

A battery on this site has a constrained value envelope. Demand charge reduction is real but the demand component is small in absolute terms. Tariff arbitrage works but the throughput is limited. Solar coordination is the strongest line because the system already exports significant midday surplus.

For sites in this category, a battery typically pays back somewhere between seven and ten years on current capex, depending on which incentives are accessible. The case improves if demand exposure is sharper than the profile suggests, or if the site is on a contestable supply with broader tariff options. The case worsens if the battery is sized larger than the site's daily cycle can absorb.

The honest answer for most sites in this category is to fix the tariff position and the active solar management first, then re-examine the battery case in twelve to eighteen months.

Site two: mid-size industrial

A mid-size industrial site running 300 kW to 500 kW peak. Operating hours wider, sometimes around the clock. Existing solar of 200 kW to 400 kW. Demand charges material on every bill. Sharp peak windows driven by specific plant cycles.

This is the strongest battery case in WA commercial. The demand charge reduction line alone often justifies the asset on a six to eight year horizon. Tariff arbitrage adds a meaningful second line. Solar coordination adds a third. Where market participation is accessible, it's a fourth.

A 200 kWh to 500 kWh battery on a site in this category typically pays back inside seven years under active management. The same battery on the same site, set to a default self-consumption mode and left alone, pays back materially later or not at all.

The difference between those two outcomes is operational, not technical. The hardware is the same.

Site three: larger commercial or campus

A larger commercial site or campus running 800 kW to 1.5 MW peak. Multiple buildings or operations. Substantial solar, often 500 kW or more. Material demand exposure. Often on a custom commercial tariff or a contestable arrangement.

The case for a battery on a site like this is rarely about whether it pays back. It usually does. The case is about sizing, integration, and which revenue lines to optimise for.

These sites are also the ones where market participation becomes material. The site is large enough to clear the eligibility thresholds, the integration cost is justifiable, and the incremental revenue from wholesale participation moves the payback inside five years on a well-sized asset.

The risk on sites in this category is overbuilding. A battery sized against a worst-case business case rather than the actual load profile underperforms its capex every year of its life.

What changes the answer

A few variables shift the case more than the rest.

Tariff structure. A site on a flat-rate energy tariff with no time-of-use signal captures none of the arbitrage line. The same site on a time-of-use tariff captures all of it. The tariff is often easier to change than the load profile. We cover this in tariff and supply optimisation.

Demand methodology. The way Western Power calculates the demand charge on a site, including which windows it considers and how it averages the readings, changes the value of a kW reduction. Two sites with apparently similar bills can have materially different demand-reduction value from the same battery.

Solar size and orientation. A solar system sized larger than midday consumption produces surplus that a battery can capture. A solar system sized close to load doesn't. East-west arrays produce a flatter generation profile that interacts with a battery differently than a north-facing array.

Load profile predictability. Battery dispatch value depends on forecasting the load and discharging at the right moment. Sites with regular, predictable load profiles are easier to model and the asset earns closer to its theoretical maximum. Sites with erratic, unpredictable load earn less from the same hardware.

The next two years of WA market changes. Network tariff restructuring, reserve capacity reform, and the AES Code are all moving the underlying maths. The case for a battery in 2027 will be different to the case in 2026, on the same site, with the same hardware. The direction of movement on most sites is toward a stronger battery case, not a weaker one. See the WA Energy Market hub for the running view.

What sinks the case

A few things genuinely kill the maths.

  • The battery is oversized for the site's daily cycle. Capex scales linearly with size. Revenue scales with how much of the capacity actually gets used. An oversized battery is paying for capacity that sits idle.
  • The site has no demand exposure. Some commercial sites genuinely have flat load profiles that don't trigger meaningful demand charges. A battery on a site like this captures only the tariff arbitrage line, which is rarely enough on its own.
  • The site is on the wrong tariff. A battery on a flat-rate tariff captures almost none of the arbitrage value. Fixing the tariff first is usually cheaper than upsizing the battery to compensate.
  • The asset is set to default mode and left alone. This is the most common reason batteries underperform in WA today. The hardware can do the job. Nobody is making the dispatch decisions.

How active management changes the numbers

Across the three site types above, active management changes the payback by between eighteen months and three years on typical configurations. On well-sized assets it can be more.

The reason is mechanical. A battery's value comes from making the right dispatch decision at the right interval against the right signals. A default-mode battery makes a generic decision that's right some of the time. An actively managed battery makes the specific decision the site requires every time.

The difference compounds across thousands of intervals a year. Across an eight-year life, it compounds to a different asset.

A commercial battery in WA is a serious capex decision. It's worth doing on more sites than were worth doing it on five years ago. It's also worth doing badly on more sites than is healthy.

The honest position

A commercial battery in WA is worth doing on more sites than it was five years ago. The market underneath is moving in favour of the case. So is the hardware cost. So is the integration ecosystem.

It's also worth doing badly on more sites than is healthy. A battery installed against a generic business case, set to default mode, and left to its own devices underperforms what the same hardware can do under active management. That's the most common outcome we see when we audit existing installs.

If a battery is on the table for your site, the right next step is a site-specific review against your actual consumption, tariff, and operational pattern. Not a brochure. Not a generic calculator. Not a quote written against an assumed load profile.

We do that as part of every site review. The review is free. The answer is sometimes yes, sometimes not yet, sometimes no. It's the answer that fits your site, not the answer that fits our pipeline.