Insights · Data Centres

Powered Land: What Property Developers Get Wrong About Data Centre Sites

July 2026 · 7 min read · Moe Abdo, Founder & Principal Advisor

Every week, another landholding is marketed as a "data centre opportunity." Most of them are nothing of the sort. The gap between land that is zoned for a data centre and land on which a data centre can actually be delivered is where fortunes are made and lost in this sector — and it is a gap that traditional property due diligence was never designed to assess.

Zoned is not deliverable

In conventional development, the site screening hierarchy is familiar: location, zoning, size, access, price. Data centre development inverts that hierarchy. The first question is not where the site is or what it costs — it is whether the site can be powered, at what capacity, and by when. A perfectly located, correctly zoned, attractively priced site with a decade-long grid connection pathway is not a data centre site. It is a paddock with good marketing.

Connection capacity is not something you can read off a zoning map. It requires engagement with the network business, an understanding of the connection queue, and a realistic view of augmentation requirements and who pays for them. It is the single largest source of both risk and value in a data centre land transaction, and it is routinely assessed last — or worse, assumed.

The four gates every site must pass

When we screen sites for data centre development, we assess four gates in order. First, power: available or securable capacity, the connection pathway, and the credibility of the timeline. Second, land use: not just current zoning but the realistic approvals pathway, including the questions councils and communities will legitimately ask about noise, water, visual impact, and construction traffic. Third, services: water for cooling where relevant, fibre routes and diversity, and drainage. Fourth — and only fourth — the conventional property questions of title, easements, contamination, and geotechnical conditions.

A site that fails gate one is out, no matter how well it scores on the others. Yet we regularly see acquisition processes that spend months on gate four before anyone has picked up the phone about gate one.

The timeline is the product

In housing development, a six-month delay is painful. In data centre development, speed-to-power is often the entire commercial proposition — the operator or hyperscaler on the other side of the transaction is buying certainty of timeline as much as land. This changes how due diligence should be structured: the objective is not simply to confirm the site is acceptable, but to build an evidenced, defensible programme from acquisition to energisation that a counterparty's investment committee will believe.

It also changes deal structure. Options, staged settlements, and conditions tied to connection milestones are not exotic in this sector — they are the appropriate response to a risk profile where the critical path runs through parties (network businesses, state agencies) that neither buyer nor seller controls.

What transfers from property development — and what does not

Developers moving into this sector bring genuinely valuable capabilities: land assembly, planning navigation, stakeholder management, and delivery discipline. What does not transfer is the assumption set. Feasibility models built on residential instincts — revenue per square metre, standard escalation, familiar consultant scopes — will systematically misprice data centre projects, usually in the optimistic direction.

The developers succeeding in this space are those who have re-learned site screening around power, rebuilt their feasibility assumptions from the sector's actual cost and programme drivers, and engaged early with the utilities and agencies that control the critical path. The discipline is the same. The hierarchy of questions is not.

The practical takeaway

If you are assessing a site — as a buyer, a landowner, or an investor — start with the power question and demand evidence, not assurances. If the answer is vague, the price should reflect it. And if you are a landowner sitting on genuinely well-located land near existing or planned network capacity, understand that its value to this sector may bear no relationship to its value under conventional highest-and-best-use analysis — but only if the connection story can be evidenced.