Your Answer Gets Graded, Not Your Site: Six Rungs From Claim to Commitment

Two directors field the same question about electric service in the same week. The sites are forty miles apart, served off the same utility's system.

The first says power is available and the utility is an outstanding partner.

The second names the substation. States its existing capacity. Notes that loads above a stated threshold trigger a system impact study, gives the study's cost and its calendar in weeks, and attaches a note from the utility's business development manager confirming all of it as of a date six weeks ago.

The second director promised nothing: no megawatts, no energization date, no cost allocation. That answer wins.

Alexandra Segers of Tochi Advisors ran this same comparison in Area Development this spring, setting "we believe power can be made available" against an answer that identifies existing substation capacity with the utility's written confirmation of timeline. Her larger finding is the one worth holding onto: a site without a clear path to power comes off the list early, whatever else it has.

The grid didn't change between those two answers. It may be literally the same grid. The difference is evidence state, and evidence state grades on a scale that runs from a claim nobody can check to a commitment somebody has signed. Tier-3 answers cluster at the bottom of that scale. Two decades of reading these responses tells me the cause is almost never a missing asset. It is a missing rubric. Nobody hands a director the grading sheet, so directors optimize for enthusiasm, which is the one variable the sheet does not score.

Here is the sheet.

Why verification cost sets the grade

The consultant works under one binding constraint: verification cost against a clock. A long list becomes a short list by way of people who cannot independently confirm anything you tell them and have no hours to try. So they sort by how expensive your claim is to check. An unverifiable claim costs infinity and gets discarded. A claim already verified by the party that controls the asset costs zero.

That is the mechanism as I read the behavior these practitioners describe. No firm publishes it as a curve. Every rung below is a position on it.

The consequence is that the grade attaches to your answer, not to your ground. A modest site with fully documented utility service outranks a superior site whose owner has not returned a call. In the 2026 State of Site Selection Pulse Check, the consultant survey the Site Selectors Guild runs with DCI and published this March, 61 percent of responding site selectors named utility and infrastructure capacity as a current site-elimination factor. Top-ranked, ahead of suitable sites at 53 percent and workforce availability and quality at 51 percent. Elimination is the operative word. Screens rank unresolved questions. They do not average your dimensions. Your effective score is your weakest answer on a dimension the prospect cares about, not your strongest answer on a dimension they mentioned in passing.

The rubric already half exists

Graduated evidence classification is not my invention. States publish versions of it, and each published tier advances on a change in evidence, not a stronger adjective.

Virginia's Business Ready Sites Program, in guidelines updated December 2025, sorts sites into five tiers.

• Tier 1. The owner is willing to market the property.
• Tier 2. Documentation committing that owner to a price and to access.
• Tier 3. Industrial zoning plus a completed diligence package, including wetlands review with Army Corps documentation and a Phase I environmental site assessment less than five years old.
• Tier 4. Infrastructure in place, or a Virginia-licensed engineer certifying that approved improvement plans are deliverable in 12 to 18 months.
• Tier 5. Construction can begin once the prospect pulls its own land-disturbance permit.

VEDP says outright that sites generally need Tier 4 or Tier 5 to be well positioned, and it makes the assessor list the work and the cost required to reach the next tier.

Michigan's MI Sites program runs Bronze to Gold on identical logic, topping out at a letter verifying utility costs and schedules. The Guild's REDI Sites designation adds a detail worth marking: its program FAQ allows outdated documents to be uploaded but warns they may receive no points, while current data scores higher. Vintage is part of the grade.

The principle, then, is settled. What I cannot find anywhere in the public record is a live client's request for information that grades responses by evidence level across power, water, workforce, site, permitting and incentives at once. Consultants describe the behavior. Nobody publishes the scale.

The six rungs below are my synthesis of documented practice and existing readiness standards. No firm has disclosed a scoring formula, and nobody will hand your response back with a number on it. Use them as an audit instrument.

The six rungs

Each rung is separated from the next by a different missing thing. That test is doing real work. Two rungs missing the same thing are one rung under two names.

Rung 1 — Claim. A favorable statement carrying no measured quantity, no named source, no date. "Ample water capacity." "Strong workforce pipeline." "Competitive incentives." Missing: any basis a stranger can verify. Checking cost is unbounded, so nobody checks, and the answer is scored as absent. You own the next move.

Rung 2 — Verified fact. A dated, attributed number or an observed condition. Permitted treatment capacity of 6.0 million gallons per day against average daily flow of 3.7, per the most recent annual report. 214 industrial maintenance credentials awarded by the community college last academic year. The parcel is zoned M-2. Missing: nobody who controls the asset has applied that fact to a deliverability question. Paper headroom is not headroom you can be served. The next move sits with the utility, the treatment authority, the landowner, the college or the permitting agency. Not with you.

Rung 3 — Authorized conditional path. The controlling party has gone on record with indicative feasibility, the assumptions underneath it, the inputs it needs from a prospect, the process, the fee, the calendar, and the conditions that would change the answer. Segers is describing this rung. It does not contain certainty. It contains bounded uncertainty, which is the material screens are built to sort. Missing: the answer has never been tested against a real project's load, chemistry, headcount or footprint, because no such project exists yet. Only the prospect supplies that.

Rung 4 — Project-specific determination. The responsible party has run the analysis for the named prospect. The utility completed a system impact study for a 42 MW load with a stated ramp. The pretreatment authority reviewed the actual process chemistry and issued a compatibility finding. The county attorney confirmed the abatement fits adopted policy at the stated investment and wage levels. Missing: nobody has accepted cost, risk or performance obligation. Advancing it takes a signature from someone with contracting authority. A governing body in open session, a utility officer executing a service agreement, the prospect countersigning.

Rung 5 — Accepted commitment. The council voted the abatement at a noticed public hearing with performance conditions and clawback written into the resolution. The prospect paid the construction deposit and signed the line-extension agreement allocating upgrade cost. The option on the parcel was exercised. Missing: the thing itself may not yet be reserved, funded, built or procured. Moving it requires the parties named in those documents to perform, plus whoever appropriates capital and issues permits.

Rung 6 — Reserved or executed. Capacity reserved on the utility's system. Transformers ordered against a slot in the manufacturer's queue. Funds appropriated to a named line item, permits issued, grading complete. Virginia's Tier 5 is a public definition of this state on the land side: utilities at the site or extendable in under 12 months, building construction able to start as soon as the prospect pulls its own land-disturbance permit. The caveats that remain concern keeping the commitment valid, not converting a promise into one.

Five disqualifiers, each distinct. No verifiable basis. No authorized path. No project-specific test. No accepted cost or risk. Nothing reserved or executed. Different problems, with different owners and different remedies.

One property of the ladder gets missed routinely. Rungs decay. A utility confirmation goes stale as load queues fill. A Phase I ages past the five-year window Virginia recognizes. A completions figure turns into a description of a cohort other employers already hired. A rung-3 answer dated 2023 grades as rung 2 today, and the REDI scoring treatment of vintage is the public confirmation that graders behave this way. Holding a rung is recurring work. It is not a climb you finish.

Where your ceiling is rung 3, and where it isn't

That line explains most of the frustration I hear from directors. They stop at rung 2 on power in the belief that rung 3 requires a live deal. It does not. Or they chase rung 4 assurances from institutions with no legal capacity to produce them, then read the refusal as indifference. Meanwhile the two dimensions where they could climb to the top of the ladder unaccompanied sit untouched.

Power. Ceiling: rung 3. Serving point, voltage, screening-level system condition, an indicative feasibility range, study steps, fees, the categories of upgrade a large load would trigger, and the utility's stated conditions for producing a project-specific answer. Unlocking rung 4 takes peak and average demand, load factor, ramp schedule, redundancy requirement, target energization date, site control, and the study fee. One structural note. A municipal utility reaches rung 3 unilaterally, because the same governing body controls both the answer and the willingness to give it. In investor-owned territory, the city has to negotiate a working partnership before anyone puts indicative numbers on paper. That asymmetry sorts communities before a prospect exists, and it does not appear anywhere on a site profile.

Water and wastewater. Ceiling: rung 3. System configuration, permitted and current flow, the local limits ordinance, pretreatment program status, review steps and calendar. Rung 4 requires gallons per day, oxygen demand and solids loading, pH, metals, fats and oils, batch discharge pattern, and the actual process chemistry. Partial readiness fools people on this dimension more than any other. Hydraulic room and chemical compatibility are two separate answers out of the same pipe. A plant with capacity to spare and no ability to accept a specific pollutant profile is not half compatible. It is a full pass for dry-process manufacturing and a full fail for wet-process categories.

Workforce. Ceiling: rung 3, and lower at the state level than most directors assume. Local community colleges can and do issue conditional letters tied to employer commitment, governance approval and funding. State customized-training programs cannot, by their own published process. Louisiana's LED FastStart describes itself as a discretionary incentive awarded to committed projects, requires at least 15 net new permanent manufacturing jobs, and builds scope only after analyzing the company's proposed operation. Georgia Quick Start's project study collects materials, technology and production sequence from company subject-matter experts before any curriculum exists. Alabama's AIDT process opens its published sequence with the company committing to create jobs in Alabama. None of the three publishes authority to promise a cohort size or a launch date for an unnamed employer. Asking them to is asking them to break their own program rules. Your ceiling is existing programs, credentials, cohort capacity, completions, instructors, equipment, eligibility rules, prior delivery experience, and a named process contact.

Permitting. Ceiling: rung 3. Current zoning, uses by right versus discretionary, the approving bodies, notice and hearing requirements, known environmental reviews, and a calendar built from your own adopted procedures rather than from optimism. Rung 4 requires exact use, emissions, discharge, hazardous materials, disturbance footprint, operating hours, construction dates.

Site. Ceiling: rung 5 or 6. Land is the one dimension where physical work and published standards let a community finish the climb with no prospect in sight. Virginia Tier 4 and 5, MI Sites Gold. Even Tier 5 leaves the land-disturbance permit to the buyer, which is the residual the standard is honest about.

Incentives. Ceiling: rung 3 in most communities, rung 5 in a few. The rung-3 version is statutory programs, eligibility thresholds, formulaic calculations, approving bodies, the meeting calendar, performance conditions and clawback structure. A pre-adopted local incentive policy with a published formula, backed by a standing appropriated fund, answers a categorically different question than an incentive requiring a council vote after the RFI lands. Directors treat incentives as inherently discretionary. The discretion is a choice about when the decision gets made.

There is a second place a community climbs past rung 3 alone. Pre-commitment funding vehicles reach a high rung on infrastructure before demand exists, by locking a project list at the moment of authorization instead of the moment of need. Georgia's special-purpose local option sales tax operates this way: the voter-approved list is fixed at referendum, so water or sewer extension money can be committed years ahead of any prospect asking. The tax mechanism is incidental. The lock is the feature.

What the upper rungs actually buy

At rung 1 you are discarded at zero cost to the consultant, and you never find out it happened. Nobody has to justify the cut. There was nothing on the page to overcome.

At rung 3, eliminating you requires an affirmative judgment that a bounded constraint is disqualifying. The Guild's Site Selection 101 presentation from September 2024 sorts screening problems into fatal flaws, serious flaws, and manageable issues that can be remedied or mitigated through negotiation with government officials. An unbounded constraint reads as fatal, because nobody can size it. A bounded constraint with an authorized conversion path reads as manageable. The burden of proof moves. That shift is the return on the preparation work.

Ben Worrell of McGuire Sponsel put the economics plainly in an August 2025 piece:

Communities that can quickly verify workforce, utility capacity, logistics and incentive readiness survive the early cuts. Rod McCants of Site Selection Group is blunter about the top end, writing in November 2025 that a site able to prove it meets a company's utility requirements often rises to the top of a short list.

My position is that the assembled proof file, not the claimed asset, is the competitive unit. What I will not claim is that a fully documented finalist becomes hard to substitute. The Guild's own process description carries a recommended site and an immediate backup through the final stage, past executive tours and reevaluation. Substitution is standard risk management all the way to the end. And no source I can find quantifies the hours a documented response saves a consultant, the probability that documentation moves a city into the finalist set, or the cost of swapping out a well-prepared community late. The switching-cost argument is an inference from a documented narrowing sequence. A reasonable inference. Not a measured one. A director building a budget request on it should know which of the two they are holding.

Two conditions for not submitting

South Dakota's state development office tells communities in its published RFI guidance not to submit when they lack a workable solution to a project's key driver, and to disclose a missing attribute rather than skip the question. That is the first stop condition and the more important one. A gap with a credible conversion path is a manageable flaw. A gap you quietly omit is a credibility loss that contaminates every other answer in the binder, the true ones included.

The second: stop when no institution will own the next action, or when the favorable version of your conversion calendar still misses the prospect's date. A rung-3 answer with a twenty-week study attached, against a prospect needing energization in fourteen months, is not a near miss. It is a no. Pursuing it burns a quarter of staff capacity that a different lane would have used better.

The ladder is a diagnostic. On any given prospect, its most valuable output may be the finding that you sit two rungs short with no path, on the one dimension the prospect ranked first.

Pull three answers you gave last year and grade them. Rung 1 or rung 2 is not fixed by a better brochure or a redesigned site profile. It is fixed by a meeting with whoever controls the fact.