Sequencing Diligence: Kill Criteria First

The scoring framework tells you how to compare sites; sequencing tells you in what order to spend your limited time and money investigating them. Getting the sequence right is where discipline pays off most, because diligence is expensive and the supply of good sites is thin, so you cannot afford to study every candidate to the same depth. The governing principle is simple and counterintuitive to beginners: spend the cheapest, fastest analysis first, and use it to eliminate sites before you spend the expensive analysis on anyone.

This is the logic of kill criteria first. Recall from Module 3 that some conditions are gates, pass-or-fail facts that no strength elsewhere can offset. The most valuable property of a good gate is that it is usually cheap to check relative to the cost of full diligence. Whether a parcel has a credible path to firm power, roughly and quickly, can often be established with a conversation with the utility, a look at the interconnection queue, and a review of nearby transmission and substation capacity, all before spending a dollar on geotechnical surveys, environmental studies, or legal work. Whether the jurisdiction has an active moratorium or hostile zoning posture can be learned from public records in an afternoon. Whether the site sits in a floodplain or an over-allocated water basin is a matter of published data. These are the questions that most often kill a site, and they are among the cheapest to answer. Answer them first.

The sequence, then, runs roughly as follows. Begin with a wide funnel of candidate regions or sites identified by starting from where power is actually available, the inversion Module 1 described. Apply the cheap gates immediately: firm power path, entitlement posture, gross water and environmental constraints, basic fiber presence. Most candidates fail one of these, and failing cheaply is a success, not a disappointment, because it preserves your resources for the survivors. Only the sites that clear every gate advance to the weighted scoring and the deeper, costlier diligence, detailed power studies, geotechnical and environmental investigation, title and legal review, formal community engagement, and negotiation of terms. By the time you are spending serious money, you are spending it only on sites that have already proven they can, in principle, be built.

The reason this ordering matters so much is the asymmetry of costs. Deep diligence on a single serious data-center site runs into significant expense and months of calendar time. Checking a kill criterion runs into hours or days and modest cost. A team that inverts the order, that falls for an attractive parcel and commissions full studies before confirming the power path, routinely discovers the disqualifying fact after it has already spent the money and, worse, burned the time. In a business where deliverable power carries multi-year lead times and good sites are scarce, time lost is often more damaging than money lost, because a competitor has secured the genuinely buildable site while you were studying a dead one. Faraday's read: the discipline of killing sites early is not caution for its own sake; it is how you stay fast in a race where the scarce asset goes to whoever confirms and commits first.

Sequencing also shapes how you engage the slow, relationship-dependent parts of the process. Because firm power is the pacing constraint, serious engagement with the utility should begin early, in parallel with the cheap gates, since the interconnection process itself takes years and a place in the queue is valuable. Community engagement deserves similar care: it is tempting to defer, but a community's posture is a gate, and early, honest engagement both tests that gate and begins building the relationship the project will need for its entire life. The mistake is to treat these long-lead relationships as late-stage paperwork; they are among the first things to initiate precisely because they take the longest to mature.

A final point on judgment within the sequence. Gates are not always crisp yes-or-no facts; sometimes a site's power path is plausible but uncertain, or a community's posture is mixed. In those cases the right move is not to force a premature kill or a premature commitment, but to identify the specific unknown and spend a targeted, proportionate amount of diligence to resolve it, before committing the full weight of studies and capital. This is where sensitivity analysis from the previous module connects to sequencing: the assumptions your ranking is most sensitive to are exactly the ones worth resolving early, because they are what will actually decide the outcome. Diligence, sequenced well, is a series of increasingly expensive bets, each one placed only after the cheaper bets have paid off. Structure it that way and you will consistently arrive at good sites faster and cheaper than teams that study everything to the same futile depth.