LocusGrid scores whether a data centre site will get through community and regulatory scrutiny. One number, 0 to 100, built entirely from the Australian public record, with every claim sourced, dated and rated for confidence.
Two real, fully sourced NSW sites. No login required during preview.
High is good. Like a credit score, a high LVI means the site is clear to proceed.
Grid and water get modelled to three decimal places. Whether the community and the regulator will let you build gets a gut feel, a consultant, and six months of finding out the hard way.
Community support now ranks second among global data centre site-selection factors, behind only power.
Transgrid states capacity to connect in Western Sydney "is largely exhausted" and that new data centres "will not be able to connect in this area without major network augmentation."
Social licence is a formal criterion in the federal approvals pathway, and the NSW Legislative Council is running a dedicated inquiry into data centres.
It is no longer scattered residents at a council meeting. In Victoria it is a structured, publicly editable wiki with a database entry for each data centre site, maintained by volunteers and campaign groups and edited continuously through August 2026.
Four pillars, 25 points each, twelve named sub-indicators. Every score traces to a public record you can open yourself. The published scorecards below were scored under rubric v0.3, versioned so two scores are only ever compared under the same method.
Sub-indicators marked in terracotta are critical: each one can end a project on its own, and is treated accordingly. See the Critical Sub-Indicator Rule below.
Anyone can add four numbers. The defensibility is in what happens at the edges, and in refusing to let a good average hide a fatal flaw.
If any pillar lands in its bottom band, the whole index is capped at 39 regardless of the raw sum. Two pillars, and it caps at 19. Real approvals do not average out: one dead grid zone or one over-allocated basin ends a project by itself, and the score has to behave the same way.
Three sub-indicators can each kill a project alone: available grid capacity, basin stress, and zoning permissibility. A severe score on any of them holds its whole pillar in the bottom band, even if the pillar's arithmetic would otherwise lift it clear. This rule exists because on our first real site, correcting one indicator would have moved the score two bands upward while the decisive fact, a transmission zone with no spare capacity, had not changed at all.
Absence of objection is not evidence of support. On an early-stage site nobody has objected yet because nobody has been asked. Where a site scores high on community licence purely because the record is empty, the report says so in plain terms rather than presenting silence as a clean bill of health. Quiet is not consent.
Every one of the twelve sub-indicators carries three things: a confidence rating, the date the underlying evidence was published, and whether the claim is primary, secondary or our own inference. A score without lineage is an opinion. This also makes a report visibly perishable, which matters: planning status, grid commitments and dam levels all move.
Our sample reports prove the method. A pilot proves it on land you are actually considering.
You nominate the parcels. We score them against the published rubric and hand back a report per site.
Pilots are priced per engagement depending on site count and complexity. Ask and you will get a number the same day, not a proposal process.
Reports are assembled by hand today, not generated by a pipeline. That is deliberate at this stage: it keeps the methodology honest and lets us calibrate against real approval outcomes before automating anything.
Both scored under the same rubric, from the same categories of public source. They land 36 points and two bands apart, which is the whole point: an index that only ever flagged risk would not be telling you anything. Below them sits the grid deep-dive, which shows how far down a single pillar goes when a customer needs the detail.
Full interactive scorecards for the AirTrunk Mamre Road campus and the Kurri Kurri data centre, including every source, confidence rating and the what-if sandbox.
Send us three parcels. In two weeks you will know which one carries the risk your grid model cannot see, and exactly why.
Only your name and email are required. Everything else helps us come back with a useful answer instead of a discovery call.