Data Center Interconnection Queue Due Diligence: How Developers Read Power Risk Before LOI
Interconnection queue diligence helps developers separate sites with real energization paths from sites with theoretical power.
Data center interconnection queue due diligence is the process of testing whether a site can realistically receive the power it needs, when it needs it, before the developer signs an LOI or commits capital. It combines utility engagement, queue analysis, transmission planning, upgrade review and schedule risk tracking.
It is now a core data center development workflow. JLL's 2026 Global Data Center Outlook says speed to power is the primary site selection criterion. Uptime Institute's 2025 Global Data Center Survey points to worsening power constraints and rising costs across the sector. A site with land, fiber and zoning upside can still fail if energization slips by 24 months.
The interconnection queue is where many of those slips first become visible.
Why queue diligence is different from a power map
A power map shows assets. Queue diligence shows congestion.
A map can tell a developer that a site is near a substation, transmission line or utility service territory. That is useful, but shallow. Queue diligence asks harder questions:
What new generation, storage and load projects are already competing for the same grid capacity?
Which upgrades are already planned, funded or delayed?
What study process applies to this specific load request?
What deposits, milestones and withdrawal penalties apply?
What is the realistic energization date after utility review?
Lawrence Berkeley National Laboratory's Queued Up 2025 data, published through OSTI, shows more than 2,600 GW of generation and storage capacity in U.S. interconnection queues at the end of 2024. That figure does not represent data center load directly. It does show the scale of grid study congestion developers are operating inside.
For data centers, the practical issue is not just whether power exists. It is whether the grid operator, transmission owner and local utility can study, approve, upgrade and serve the load inside the project schedule.
The workflow before LOI
A disciplined queue diligence workflow has six steps.
1. Define the load request by phase
Do not start with the final campus number alone. Start with phased demand.
A 300 MW campus may need 48 MW for phase one, 96 MW by year two and the balance over several years. That phasing changes utility options. It may also create interim solutions, such as temporary service, staged feeders, behind-the-meter generation or battery support.
AI can help normalize load cases across sites. It can convert equipment plans, customer requirements and development phasing into a structured load request. The final load letter still needs engineering review.
2. Identify the serving utility and relevant grid operator
The same county can contain multiple utilities, transmission owners and planning processes. A developer should identify the serving utility, transmission owner, balancing authority, ISO or RTO exposure and any state-level commission process.
This matters because the timeline is procedural. A site in PJM, ERCOT, MISO or a vertically integrated utility territory will not follow the same study path.
3. Read visible queue pressure
Public queue data is imperfect, but it is still useful. Developers should review generation queues, storage queues, transmission planning dockets, utility load forecasts and large-load tariff filings.
The question is not 'how many projects are nearby?' The question is 'which projects could consume upgrade capacity, trigger network reinforcements or force restudies?' A nearby battery project, solar project or hyperscale load request can change the answer.
AI is useful because queue data changes often and uses inconsistent naming. An agent can monitor queue updates, match projects to grid nodes, flag withdrawals and identify projects that share likely constraints.
4. Separate study risk from construction risk
Interconnection diligence often blends two different risks.
Study risk is the time required to get a utility answer. Construction risk is the time required to build the upgrades after that answer exists.
Both matter. A study delay can push the investment decision. A transformer or breaker delay can push energization. The diligence memo should track them separately.
5. Price upgrade exposure
Some sites require shallow work. Others trigger major network upgrades. Before LOI, developers rarely have final cost certainty, but they can bracket exposure.
The diligence team should identify likely categories: new feeders, substation expansion, transformer additions, protection upgrades, transmission reconductoring, new switching stations, easements or customer-funded dedicated facilities.
Each category should be assigned a schedule band, cost range and confidence level. That is enough to decide whether the site deserves deeper pursuit.
6. Create a milestone tracker
Queue diligence should produce a live tracker, not a one-time memo. The tracker should include utility contacts, study applications, deposits, data requests, feeder route decisions, easement dependencies, expected study dates and board or commission approvals.
Good AI makes this tracker useful. It watches utility filings, public agendas, queue updates, project documents and internal notes. It escalates changes that affect the energization date.
What AI can and cannot do
AI can assemble evidence faster than a human team can manually search for it. It can compare queues across markets, summarize utility filings, pull transmission planning language and flag missing assumptions.
AI cannot approve service. It cannot negotiate utility obligations. It cannot certify engineering feasibility.
The best workflow is human-in-the-loop. AI builds the evidence file and keeps it current. Utility specialists and engineers make the call.
The decision rule
Before LOI, the team should be able to answer four questions:
What is the earliest credible energization date?
What assumptions does that date depend on?
What upgrades are likely and who pays?
What new information would make the site a no-go?
If those answers are vague, the site is not ready for capital committee. It may still be worth pursuing, but the risk should be priced as unknown.
Power risk is not binary. It is a schedule, cost and confidence problem. Queue diligence turns that risk into something developers can underwrite.