Corrected 21 October 2026. From 3 August this page’s standfirst and descriptions said the newest month in this dataset undercounts by about 78 per cent, and from 3 October by 78 to 88 per cent. Those figures are the rises from a month’s first count to the next edition’s (May 350 to 624, June 420 to 779, July 411 to 771). Measured against the later figure, the first counts were 44, 46 and 47 per cent short. The figures in the notes and body below are right as rises, including the three the 3 October note calls the undercount; the standfirst now gives the shortfall.
Updated 21 October 2026. The Regulator published a fifth edition of the postcode data, stamped 30 September 2026, on 20 October (landing page “Last updated 20 October 2026”). We downloaded and summed it on 21 October over the same 29 postcodes by the same method, and the headline and standfirst now carry it. The first year restates to 11,781 systems and 306,229 kWh, and June now reads 895. August 2026, the newest month on 3 October, went from 457 to 795, a rise of 74.0 per cent, so its first count was 42.5 per cent short. That fourth measurement is the smallest yet: as a rise the range is now 74 to 88 per cent, as a shortfall 43 to 47. The charts and postcode table further down still show the 30 June edition, for the reason the 3 October note gives.
| First year (Jul 2025 to Jun 2026), then later months | 31 Aug edition | 30 Sep edition |
|---|---|---|
| Battery systems | 11,677 | 11,781 |
| Usable capacity, kWh | 303,766 | 306,229 |
| Average system, kWh | 26.0 | 26.0 |
| April 2026 (peak) | 1,894 | 1,914 |
| June 2026 | 841 | 895 |
| July 2026 | 771 | 879 |
| August 2026 | 457 | 795 |
| Top ten postcodes, first year | 31 Aug edition | 30 Sep edition |
|---|---|---|
| 2320 Maitland area | 999 | 1,006 |
| 2287 | 760 | 766 |
| 2285 | 742 | 749 |
| 2324 | 670 | 679 |
| 2322 | 672 | 678 |
| 2325 Cessnock | 666 | 674 |
| 2290 | 663 | 670 |
| 2321 | 628 | 637 |
| 2283 | 626 | 632 |
| 2280 Lake Macquarie | 626 | 631 |
The same ten postcodes again. 2324 is back above 2322, by one install, and 2283 now leads 2280 by one, where on 3 October they tied.
Updated 3 October 2026. The Regulator has published a fourth edition of the postcode data, stamped 31 August 2026 (landing page “Last updated 17 September 2026”), which we downloaded and summed on 3 October over the same 29 postcodes and by the same method as every edition before it, from the installations and capacity workbooks. The headline and standfirst now carry it. On this edition the first year, July 2025 to June 2026, restates to 11,677 systems and 303,766 kWh, and June, which printed 420 when this page was written and 779 at the last update, now reads 841. July 2026, the newest month at the last update, went from 411 to 771, a rise of 87.6 per cent, so the newest-month undercount has now been measured three times, at 78.3, 85.5 and 87.6 per cent. Both charts and the postcode table below still show the 30 June edition and are left as published: the first chart is the series as that edition recorded it, the second is the May-to-June revision it measured, and overwriting either would erase the comparison this page exists to make. The tables here set the editions side by side.
| First year, July 2025 to June 2026 | 30 June edition (as published) | 31 July edition (18 Aug note) | 31 August edition |
|---|---|---|---|
| Battery systems | 10,942 | 11,494 | 11,677 |
| Usable capacity, kWh | 282,500 | 298,973 | 303,766 |
| Average system, kWh | 25.8 | 26.0 | 26.0 |
| April 2026 (peak month) | 1,751 | 1,851 | 1,894 |
| June 2026 | 420 | 779 | 841 |
| July 2026 (outside the year) | not yet published | 411 | 771 |
| Top ten postcodes, first year | 30 June edition (as published) | 31 August edition |
|---|---|---|
| 2320 Maitland area | 938 | 999 |
| 2287 | 713 | 760 |
| 2285 | 696 | 742 |
| 2322 | 630 | 672 |
| 2324 | 637 | 670 |
| 2325 Cessnock | 624 | 666 |
| 2290 | 628 | 663 |
| 2321 | 589 | 628 |
| 2280 Lake Macquarie | 602 | 626 |
| 2283 | 581 | 626 |
The ten postcodes are the same ten; three adjacent pairs have swapped places (2322 above 2324, 2325 above 2290, 2321 above 2280) and 2280 and 2283 now tie at 626. The year-window figures for each postcode are the twelve monthly columns from July 2025 to June 2026 summed, not the workbook’s total column, which now runs to August 2026.
Updated 18 August 2026: the prediction held, and June was not a crash. This page said the newest month in the Regulator’s file undercounts by about 78 per cent, measured from a single earlier revision. The Regulator has now published a third edition, current to 31 July 2026, so that estimate can be tested rather than trusted. June, which read 420 when this was written, now reads 779, a rise of 85.5 per cent. The decay curve held a second time: one month back rose 9.6 per cent, two months back 5.7 per cent, three months back 1.3 per cent, and everything older moved by under 1 per cent. On the new file the first year restates to 11,494 systems and 298,973 kWh, up from the 10,942 and 282,500 kWh below, and the newest month is now July 2026 at 411. If July behaves like the two revisions we have measured it lands somewhere around 730 to 760, which is our arithmetic on two observations rather than a forecast. Everything below is the year as the June edition recorded it and is left as published, because the point of this page is the comparison and rewriting the base figures would erase it.
In July this masthead charted the first eleven months of the scheme from the same source, and warned in passing that the latest month always looks smaller than it turns out to be. That was the right caution, but it was only a caution. The Regulator has since published its June edition, and because the earlier figures were recorded at the time, the two files can be set against each other. That turns the warning into a measurement, and the measurement is larger than the warning implied.
The year, month by month
Validated battery installations per month, summed across the Hunter postcode set listed in the methodology note. Source: Clean Energy Regulator SRES postcode data, installations workbook, SGU-Battery sheet, data as at 30 June 2026.
The shape is a scheme that took most of a year to find its stride. From 842 systems in the opening month the pace wandered through spring, dipped over January to 760, then climbed steeply: 1,033 in February, 1,324 in March and 1,751 in April, more than double the opening month. Then the line appears to fall off a cliff. It is the cliff that is not real.
What two editions of one file reveal
The Regulator says as much itself, in a note printed at the top of every sheet in the
workbook: renewable energy target legislation allows a 12 month creation period for
registered persons to create their certificates
, and so the 2025 and 2026 figures
will continue to rise
. An installer who fits a battery in June has until the
following June to lodge the paperwork that puts it in this dataset. Recent months are
therefore always partial.
What the Regulator does not publish is how partial. Comparing its 30 June edition against the 31 May one answers that for every month at once, because each month appears in both files and any difference is late paperwork arriving.
Percentage increase in the Hunter battery count for each month, measured between the workbook current to 31 May 2026 and the workbook current to 30 June 2026. Our calculation from the two editions; the earlier figures are those this masthead published on 18 July 2026.
The decay is orderly. The month that was newest when we last looked, May, has grown from 350 to 624, a rise of 78.3 per cent in about eight weeks. One month further back, April rose 6.8 per cent, which is why its 1,640 is now 1,751. Two months back, March moved 2.7 per cent. By three months back the correction is 1.4 per cent, and beyond four months nothing shifts by more than half a per cent. In practical terms this dataset takes roughly three months to settle, and only the last two months are seriously incomplete.
Applying that to June’s printed 420 is straightforward arithmetic: if it fills in as May did, it lands somewhere near 750. We would not call that a forecast. It is one observed revision cycle rather than a model, and a single cycle can be atypical. But it is grounded enough to say what June is not, which is a collapse in demand. The honest summary of the year is that installs were still climbing through autumn, and the last two bars cannot yet be read.
Where the batteries went
| Postcode | Systems | kWh | Avg kWh |
|---|---|---|---|
| 2320 Maitland area | 938 | 24,722 | 26.4 |
| 2287 | 713 | 18,437 | 25.9 |
| 2285 | 696 | 18,186 | 26.1 |
| 2324 | 637 | 14,970 | 23.5 |
| 2322 | 630 | 16,376 | 26.0 |
| 2290 | 628 | 15,385 | 24.5 |
| 2325 Cessnock | 624 | 16,595 | 26.6 |
| 2280 Lake Macquarie | 602 | 15,324 | 25.5 |
| 2321 | 589 | 15,689 | 26.6 |
| 2283 | 581 | 14,668 | 25.2 |
Maitland’s 2320 leads the region again, as it did on every measure in the earlier piece, and its lead has widened from 877 systems to 938. The spread across the top ten is narrow, which is the more interesting fact: nine postcodes sit between 581 and 713, so this is a region-wide pattern rather than a couple of enthusiastic suburbs. Average system size barely varies, from 23.5 kWh to 26.6 kWh, clustering near the 25.8 kWh Hunter-wide average. That consistency suggests households are mostly buying a standard product rather than sizing to their own consumption.
For context, the same workbooks count 97,894 rooftop solar systems and 696,822 kW across those postcodes since 2001. A year of batteries has reached the equivalent of roughly one in nine of those solar households.
How we did this
Both Clean Energy Regulator workbooks, capacity and installations, were downloaded on 3 August 2026 and summed with a script over a fixed set of 29 Hunter postcodes: 2300; 2280 to 2287, 2289 and 2290; 2318 to 2327; 2328, 2329; 2330; 2333; and 2334 to 2337. The set is deliberately identical to the one used in July, so the two editions are comparable. Battery figures come from the SGU-Battery sheets, which begin in July 2025 because that is when the scheme began.
One trap worth recording for anyone repeating this: the grand total is the
last column headed “Total Installation Quantity”. A column headed
Historic Total Installation Quantity (2001 - 2010)
sits near the start of
each sheet and matches a naive search, but covers only the first decade.
The revision figures are our own calculation, comparing each month’s value in the 30 June workbook against the value this masthead published on 18 July 2026 from the 31 May workbook. Those earlier figures were independently re-derived from the source file on 21 July 2026 before being relied on here.
What we have not done. We have not asked the Regulator whether the lag differs by state, technology or installer size, so the pattern described is the Hunter’s and one cycle of it. We have not adjusted any published figure upward; every number above is as printed in the source. Postcodes 2331 and 2332 have no rows in either workbook, which we have not been able to explain and have not counted.
Sources
Production note. Produced with AI under human editorial direction, reading the documents themselves, from the 6 primary sources below. How this masthead works.
- Clean Energy Regulator, SRES postcode data, installations and capacity workbooks, fifth edition, data as at 30 September 2026 (landing page reading “Last updated 20 October 2026” and “current as at 30 September 2026”; both workbooks downloaded and totalled 21 October 2026). The figures in the headline, standfirst and the 21 October note: 11,781 systems and 306,229 kWh for July 2025 to June 2026, June 2026 at 895, August 2026 at 795, and the first-year top ten. Summed over the same 29-postcode set; the fifteen monthly columns reconcile to the workbooks’ own total columns exactly (13,927 systems, 357,696 kWh).
- Clean Energy Regulator, SRES postcode data, installations and capacity workbooks, fourth edition, data as at 31 August 2026 (landing page then reading “Last updated 17 September 2026” and “current as at 31 August 2026”; both workbooks downloaded and totalled 3 October 2026). The figures in the 3 October note: 11,677 systems and 303,766 kWh for July 2025 to June 2026, June 2026 at 841, July 2026 at 771, and the first-year top ten. Summed over the same 29-postcode set; the fourteen monthly columns reconcile to the workbooks’ own total columns exactly (12,905 systems, 332,645 kWh).
- Clean Energy Regulator, Small-scale installation postcode data: the landing page, stating the data is current as at 30 June 2026 and last updated 23 July 2026. Read 3 August 2026.
- Clean Energy Regulator, SRES postcode data, installations and capacity workbooks, third edition, data as at 31 July 2026 (landing page then reading “Last updated 13 August 2026”; both workbooks downloaded and totalled 18 August 2026). The revised monthly series behind the update at the top of this page, the restated first-year totals of 11,494 systems and 298,973 kWh, and July 2026 at 411. Summed over the same 29-postcode set, and the thirteen monthly columns reconcile to the workbooks’ own total columns exactly.
- Clean Energy Regulator, SRES postcode data, installations: the SGU-Battery and SGU-Solar sheets, the monthly series, the postcode counts, and the 12-month certificate-creation note quoted above. Downloaded 3 August 2026.
- Clean Energy Regulator, SRES postcode data, capacity: usable battery capacity in kWh and solar rated output in kW. Downloaded 3 August 2026.
- The Connection Point, Eleven months, 10,072 home batteries: the 31 May edition figures against which the revisions here are measured.
Install batteries in the Hunter, or read this data differently? Tell us and we will check it against the workbooks and log the outcome here.