Australian grid-connection intelligence
GridFlex calculates realistic curtailable capacity at specific Australian connection points, so developers can negotiate faster power and networks can assess flexible-demand proposals with a defensible method — not a guess.
Illustrative diagram. Solid lines are confirmed. Dashed lines are modelled or indicative — and labelled as such, every time.
The problem
Proximity to a transmission line has never meant deliverable capacity. Now the gap between the two is the single biggest obstacle standing between a data-centre project and an energisation date.
"There is insufficient capacity in Sydney's transmission network today to connect any of the 20 GW of connection enquiries we have received for new large electrical loads." — Transgrid, EGM Network, public statement, August 2026
What we do
GridFlex answers a single question — how much flexible capacity genuinely exists at a given connection point, and what would a data centre have to agree to in order to use it — for the two parties who need that answer to actually negotiate a connection.
A commissioned study of one or more real connection points: estimated firm capacity, estimated flexible capacity on top, expected curtailment frequency and depth, and what that pathway is worth against waiting for a conventional firm connection.
Regulators are moving toward assessing large loads on their flexible profile rather than nameplate capacity — but no standard method exists yet for doing that. GridFlex's methodology is built to become that assessment engine as flexible connection agreements are formalised.
A worked example
Net present value in A$ millions for a 50 MW data centre, grid-only configuration. Three pathways that start earlier come out at roughly double the value of waiting for a firm connection.
| NPV, A$m | Grid only 159 A$/MWh | Grid + battery 131 A$/MWh | Grid + solar + battery 129 A$/MWh | Gas primary 154 A$/MWh | Gas island first 162 A$/MWh |
|---|---|---|---|---|---|
| Firm | 106.3 | 106.3 | 106.3 | 208.6 | 182.7 |
| Staged | 219.5+113.2 | 219.5+113.2 | 219.5+113.2 | – | – |
| Non-firm (N rating) | 220.0+113.7 | 220.0+113.7 | 220.0+113.7 | – | – |
| Flexible only | 217.7+111.4 | 217.7+111.4 | 217.7+111.4 | – | – |
Firm = 62.5 MW at 4.5 years on grid, 2 years on gas. – = not modelled. Illustrative model output from public ratings and stated assumptions; revenue and connection-cost inputs are assumptions. Not a network commitment.
The confidence model
A capacity figure is only useful if you know how much to trust it. GridFlex grades every output on a seven-level evidence scale — from a public data point to a signed connection agreement — and never presents one as the other.
Who it's for
A flexibility commitment only works if you control the load. That fits AI-factory operators and neoclouds running their own compute — not colocation providers whose tenants hold the uptime guarantees.
The gap we work in sits at transmission scale — the large connection points serving 100 MW+ campuses — where no standard flexible-connection product exists yet, unlike some distribution-level offerings already in market.
Why now
Where we are
GridFlex is a new Australian company building this methodology from public AEMO data, starting with real, named connection points rather than a broad national platform. We are not a network operator, an engineering consultancy, a demand-response operator, or a guarantee of available capacity — and we say so before anyone has to ask.
Get in touch
Whether you're negotiating a connection, assessing a flexible-demand proposal, or working on the research side of this problem, we'd like to hear from you.
Email us directly
Based in Adelaide, working across the NEM. You can also message the GridFlex page on LinkedIn.