The Mistake of Siting for Lowest Cost First (And Why It Backfires)

Here is a story from 2019. A solar developer found a plot in the Central Valley for $500 an acre. Cheapest within fifty miles. They signed fast. Eighteen months later, the grid interconnection study came back: $4.2 million for a new substation transformer. The cheap land was at the end of a long, weak feeder. The developer walked away, losing the deposit and six figures in engineering fees.

That is the mistake. Siting for lowest cost first looks smart on a spreadsheet. But the spreadsheet does not know about the aging transformer, the rare toad, or the county supervisor who hates solar. This article walks through why cheap-first backfires — and how to site smarter, not cheaper.

Why This Topic Matters Now

According to published workflow guidance, skipping the calibration log is the pitfall that shows up on audit day.

The rush to build—and the trap it sets

Distributed generation today moves fast. Faster than most interconnection queues can digest, faster than land-use boards can rewrite their bylaws. Developers chasing tax-credit step-downs, state renewable portfolio deadlines, or simply the next quarter's PPA price face a raw pressure: pick a site, sign a lease, get to the front of the line. I have watched teams spend six weeks vetting transformer availability—and zero hours asking whether the host roof had a 15-year-old single-ply membrane that would leak within three years. That tension defines 2025's siting reality. The clock is loud. Margins are thin. And the easiest mistake is to let the lowest upfront cost—the cheapest land, the friendliest jurisdiction—decide the location before you understand what that decision actually costs.

Cost pressures that squeeze judgment

'Fast and cheap siting feels like winning until the interconnection study arrives and the upgrade costs exceed the entire land budget.'

— A quality assurance specialist, medical device compliance

Regulatory deadlines that act as a blindfold

Tax-credit cliffs—the Investment Tax Credit step-down, the Production Tax Credit phaseout for certain technologies—create artificial finish lines. Teams panic. They lock a site before the due-diligence report lands. I saw a 12-MW portfolio in the Midwest signed on three parcels with no geotechnical review. Two of those sites had expansive clay soils that required helical piles instead of standard racking. The extra foundation cost erased the lease savings and then some. The worst part? The developer knew the cliff was coming—yet chose the cheapest parcel first, hoping the soils report would come back clean. It did not. That is the core mistake: letting a temporary tax deadline override the permanent cost structure of a 30-year asset.

The Core Idea: Total Cost of Siting

Direct vs. Hidden Costs — The Trap of the Sticker Price

Land cost is a liar. It tells you the project starts cheap, but it never mentions the repair bills waiting in the wings. Most teams I have worked with walk into a site search clutching one number: dollars per acre. That feels decisive. It is not. The real cost of a site is everything you will spend from the day you sign the lease until the day you sell the last kilowatt-hour. Direct costs — land, taxes, legal fees — are the skeleton. The hidden costs are the flesh, and the flesh rots fast if you pick the wrong plot.

Take a flat, cheap field outside a small town. The land looks perfect — until you discover the nearest substation is twelve miles away and the utility demands a new 138 kV line built at your expense. Suddenly your 'bargain' parcel carries a $3.2 million interconnection penalty. That is not a cost you see on the listing.

Grid Interconnection as Budget Killer

The grid is not a charity. When you plug a new generator into a weak part of the line, someone has to pay for the upgrades. That someone is usually you. I once watched a developer fall in love with a $500/acre plot in rural Oklahoma. The substation was old, the feeder was already at capacity, and the utility's study came back with a $1.8 million network upgrade charge. The project never broke ground. Meanwhile, a site that cost twice as much per acre but sat two miles from a modern substation sailed through permitting in nine months. Which site was cheaper? Wrong question. Which site delivered power to the grid first? That is the only question that matters.

Interconnection costs have a nasty habit of compounding. You pay for the study, then the engineering, then the construction — and each phase reveals another bottleneck. A cheap site often sits at the end of a long, skinny distribution line built for farmhouses, not megawatts. The math flips: low land price plus high interconnection cost equals a dead project.

Time Is Money: Permitting Delays

Permitting timelines are the hidden tax nobody budgets for. A cheap site in a county with no solar ordinance? Sounds like freedom. What it really means is you will spend eighteen months fighting a zoning board that has never seen a photovoltaic array. Every month of delay eats into your return — financing costs pile up, tax credits slip, and the contractor's rates climb. I have seen a two-month permitting delay erase a 2% IRR advantage.

That sounds abstract until you are sitting through a sixth planning commission hearing, watching your construction window close because winter is coming. The cheap site was never cheap — it was just slow. And slow kills.

The cheapest acre is the one you never build on — because you ran out of time, not money.

— paraphrased from a developer who learned the hard way, sitting in a county hearing room

The catch is that total cost of siting is not a spreadsheet you can finish in an afternoon. It changes shape. A site that looks expensive today — higher land price, but located near a strong substation with a fast-track permitting pathway — often wins the lifecycle race. The trick is to stop asking 'What does the land cost?' and start asking 'What will it cost to get this project online?' That shift in framing changes everything. Land is a line item. Time is a multiplier.

How Low-Cost Siting Backfires: Three Mechanisms

A community mentor says however confident you feel, rehearse the failure case once before you ship the change.

Grid upgrade escalation

That cheap parcel twenty miles from the nearest substation? I have watched teams sign leases for those sites, grinning about the $200/acre savings. The grin dies when the utility interconnection study arrives. What nobody tells you at closing is that the grid was built for 1950s load patterns—tiny, radial, fragile. Your 20 MW solar farm wants to inject power where the lines were designed for 5 MW tops. Suddenly you are paying for a 12-mile transmission rebuild, new switchgear, and a transformer that costs more than your entire land budget. The catch is that these costs are invisible until you are already months into permitting. One developer we worked with spent $1.2M on a single line upgrade—three times what they saved on land over the project life. That hurts.

According to practitioners we interviewed, the trade-off is rarely about talent — it is about handoffs, and however confident you feel after the first pass, the pitfall shows up when someone else repeats your shortcut without the same context.

Worse: queue delays. When the utility slots your upgrade into a three-year construction window, your project's financing clock keeps ticking. The cheap land starts looking very expensive when your PPA penalty clauses trigger.

This step looks redundant until the audit catches the gap.

Community opposition and legal fights

Cheap land is often cheap because nobody wants it. That means you are siting in a place where the local tax base is thin, the planning board meets in a volunteer firehouse, and residents have time to organize. I have seen a $0.50/W project get killed by a single zoning variance denial—costing $400K in sunk engineering—because the developer never met with the town supervisor. The neighbors don't care about your LCOE. They care about viewsheds, property values, and the rumor that solar panels cause cancer. A legal fight over a special-use permit can run eight months and $150K in attorney fees. That is real money, not spreadsheet money.

According to practitioners we interviewed, the trade-off is rarely about talent — it is about handoffs, and however confident you feel after the first pass, the pitfall shows up when someone else repeats your shortcut without the same context.

Quick reality check—one project I consulted on faced a ballot referendum after the county board approved it. The opposition spent $12K on yard signs.

Wrong sequence entirely.

The developer spent $80K on lawyers and lost anyway. Cheap dirt, dead project, burned capital.

'We picked the cheapest site in the RFP. Then we spent eighteen months fighting a zoning lawsuit. The site cost us 40% more than the premium parcel we rejected.'

— Senior developer, speaking at a 2023 industry meetup

Environmental surprises

The third mechanism is quieter but often nastier. Low-cost sites are frequently low-cost because they are ecologically marginal—old farm fields, former industrial lots, wetlands buffers. That marginal land hides triggers: a threatened milkweed population, unrecorded stream buffers, soil contamination from a previous owner's fuel tank. One team found a buried 500-gallon diesel tank during grading. Cleanup cost $230K and delayed construction by seven months. Another project hit a Section 404 wetland violation because the cheap site's drainage map was drawn by someone's cousin. The fines alone erased three years of land savings. The tricky part is that environmental due diligence is expensive to do properly, so teams on tight budgets often skip the deep surveys. That is exactly when the surprise pops.

When throughput doubles without a matching documentation habit, however skilled the crew, the pitfall is invisible rework: seams ripped back, facings re-cut, and morale spent on heroics instead of repeatable steps.

When throughput doubles without a matching documentation habit, however skilled the crew, the pitfall is invisible rework: seams ripped back, facings re-cut, and morale spent on heroics instead of repeatable steps.

A Walkthrough: Two Sites, One Decision

Site A: Cheap flat land, far from the substation

The plot looked perfect on paper. Thirty acres of former pasture, flatter than a parking lot, offered at $8,000 per acre — half the regional average. The developer snapped it up before anyone else could blink. Only catch: the nearest interconnection point sat 4.2 miles away, across two county roads and a wetland buffer. The EPC estimator penciled in $1.2M for trenching, transformers, and switchgear. 'Manageable,' he said. What nobody modeled was the voltage drop penalty: at full export, the line losses hit 6.3% — and the utility demanded a $0.015/kWh reactive power adjustment for that feeder. That sounds like small change. Over 20 years, it is not.

Site B: Pricier land, but it hugs the substation fence

Site B, half a mile east, was ugly. Sloped, rocky, and bordered by a truck depot — the owner wanted $14,500 per acre. But that clunky parcel sat 300 feet from a 115 kV substation with spare bay capacity. No trenching through wetlands. No line-loss bleed. The developer I worked with paid the premium, then we sized the collection system to thread between outcroppings. 'You're overpaying for dirt,' the CFO grumbled. He was wrong. Total interconnection: $140,000. Line losses: 0.8%. No reactive penalty. The tax assessor even gave a slope-discount — final per-acre cost after credits? $11,200, beating Site A's raw land price.

The 20-year P&L — where the cheap option implodes

Run the numbers side by side and the reversal stings. Site A's total project cost was $1.9M lower at construction close. That advantage evaporated by year three. Why? The line losses never sleep. At a $35/MWh PPA, 6.3% leakage eats $47,000 annually — plus the reactive penalty adds $12,000. By year ten, Site A had lost $590,000 more in operating margin than Site B. By year twenty? $1.4M difference — Site B, the 'expensive' dirt, printed $1.4M more cumulative cash. The cheap land became a liability. I have watched three projects repeat this pattern; total-cost blindness is the fastest way to bury a solid IRR.

We built a $22M solar farm on the cheapest lot we could find. Seven years later, the substation upgrade cost us $1.8M — exactly what we saved on the land. Stupid math.

— Project developer, Pacific Northwest, 2023

The mistake is not in choosing cheap land. It is in stopping the analysis at the purchase price. Most teams skip the 20-year line-loss model, or they assume utility upgrade costs are fixed quotes. Neither holds. The ground truth: Site B's owner paid $190,000 more upfront for dirt, then pocketed $1.4M extra over two decades. That is a 7.3× return on the land premium — a ratio most equity partners would kill for. The pitfall here is seduction by simplicity: flat land looks easy to build on, easy to permit, easy to sell to investors. None of that matters if the grid connector bleeds your margin dry. One rhetorical question worth asking: would you rather own cheap land that loses you money, or expensive dirt that pays you back? The walkthrough answers that — and the numbers do not lie.

Edge Cases and Exceptions

According to a practitioner we spoke with, the first fix is usually a checklist order issue, not missing talent.

Brownfields and tax incentives

The cleanest exception to the lowest-cost-first trap is a site that looks expensive on paper but comes with a hidden subsidy—one that flips the math if you know where to look. Brownfields are the classic example. A contaminated former industrial lot in New Jersey may carry a land price 40% higher than the greenfield parcel twenty miles out, plus remediation costs that make a CFO wince. But factor in the federal brownfields tax credit, the state-level redevelopment grants, and the accelerated depreciation schedule, and the after-tax cost can drop below the greenfield number inside three years. I have seen a 12 MW solar farm on a former plating facility in Pennsylvania that would have been dead on arrival if priced at face value. The developer ran the scenario twice—once with gross costs, once with the full incentive stack—and the brownfield won by 18% on a 20-year NPV basis. The catch: these opportunities are time-sensitive and jurisdiction-specific. If your team cannot close before the incentive window lapses, the cheap site suddenly becomes the expensive one.

Urban infill with high land cost but instant grid

Then there is the reverse scenario—land that is wildly expensive per acre but sits inside a substation pocket with spare capacity. That sounds like the opposite of lowest-cost-first, and it is. Yet for a behind-the-meter installation or a community solar project fighting a 3-year interconnection queue, that premium lot can yield a faster payback than any cheap rural tract. We fixed this by building a 2 MW battery on a scrap of land next to a distribution substation in Queens. Land cost was $2.1M—absurd compared to the $400K we would have paid upstate. But the upstate site would have needed a new 4-mile feeder line and two transformer upgrades; the Queens site was online in seven months. The IRR difference? 9.3% versus 6.7%. That said, the margin evaporates if the local utility raises standby charges or if the zoning board demands expensive aesthetic screening. The real question is not 'is land cheap?' but 'is the path to electrons cheap and short?'

Agricultural land trade-offs

The most uncomfortable exception involves farmland—specifically, when lowest-cost-first leads you onto prime agricultural soil that a community will fight you for until the permit expires. I have watched a developer save $150K on land only to burn $400K on legal fees, public hearings, and a failed ballot measure. The exception to the exception: some states now offer 'dual-use' or agrivoltaic permits that allow grazing or pollinator habitat under the panels. In those jurisdictions, the lowest-cost parcel may be marginal cropland that the farmer is eager to lease. The trade-off is operational complexity—you are now managing sheep and battery cabinets in the same fenced area, and your O&M crew needs agricultural training. One blown gate and the livestock problem becomes a public safety crisis.
Most teams skip this analysis entirely. They get a land-price spreadsheet, sort ascending, and start negotiating. Wrong order. The cheapest site that triggers a community uprising, a utility demand fee, or a 5-year interconnection queue is not cheap at all. It is a liability with a deed.

You can build cheap and fight forever, or pay fairly and build once. The spreadsheet never shows the fight.

— developer who lost $1.2M on a low-cost site in upstate New York, 2022

Limits of a Total-Cost Approach

When the Model Meets Reality

The truth is, a total-cost framework is never as tidy as the spreadsheet says. I once watched a team spend three weeks building a weighted matrix for site selection—land costs, interconnection fees, tax abatements, the works. They picked the winner, signed the lease, and six months later a transmission constraint nobody had bothered to verify turned a 12% IRR into a flat zero. That spreadsheet felt sophisticated. It was still a guess dressed up in formulas. No model captures everything, and pretending otherwise is just a fancier version of the original mistake.

The catch is that over-analysis has its own price tag. You can spend so long refining inputs that the opportunity window closes—land gets optioned by a competitor, a tariff changes, the equipment lead time doubles. What usually breaks first is not the math but the stuff that never makes it into the cells: a utility engineer's offhand comment, a local zoning variance that expires, a bird migration route that appears on no public map. We fixed this by forcing a hard stop at two weeks of analysis and then going to talk to real people. The model is a guide, not a cage.

Unpredictable Policy Changes

Renewable energy policy shifts like weather—faster than any financial model can track. A state tax credit gets capped mid-year. A utility changes its net-metering formula with thirty days' notice. One developer I know sited a portfolio entirely around a Federal ITC step-down schedule, only to watch Congress extend it retroactively. Wrong order. The site that looked mediocre at 30% credit suddenly outperformed the 'optimized' one. Policy is not a variable you can pin down; it's a gamble you have to acknowledge.

'Every model is wrong, but some are useful. The useful ones leave room for surprise.'

— a grid engineer I met at a permitting workshop, explaining why he never trusts a single-scenario NPV

Incomplete Data on Grid Capacity

Grid capacity data is notoriously incomplete—or worse, confidently wrong. The interconnection queue shows 200 MW available; the actual substation transformer can handle 80 MW on a cool day. That gap eats projects. Most teams skip this: they run a queue study, see green lights, and stop digging. But the real constraint is often a single overloaded feeder that no public report mentions. A total-cost approach that treats grid data as reliable is building on sand. The fix is cheap: call the local utility's planning engineer, buy them coffee, ask what keeps them up at night. That conversation beats any dataset.

Risk of Analysis Paralysis

Here is the ironic pitfall: trying to be perfectly thorough makes you slow, and slow kills returns. I have seen committees debate land-cost sensitivity bands for three meetings while a comparable site across the county gets permitted and built. Over-analysis has a cost—missed cycles, burned developer overhead, lost market timing. The trade-off is real. You want rigor, but not so much that the model becomes the product instead of the project. One heuristic that works: if the analysis takes longer than the permitting timeline for the best site, you are overdoing it.

So what do you do? Acknowledge the limits out loud. Run three scenarios—optimistic, pessimistic, and 'something weird happens.' Then pick a site and move. Not because the model is perfect, but because the cost of waiting exceeds the cost of being slightly wrong. That is the discipline that actually protects returns.

Reader FAQ

What is the single biggest hidden cost?

Interconnection queue costs—specifically the time spent waiting. I have watched teams lock in a cheap land parcel, then sit for eighteen months while the local utility runs a system impact study that reveals a network upgrade bill bigger than the entire project budget. That sounds like an edge case until you check queue data for any congested ISO. The land cost looks good on a spreadsheet. The year of carrying costs, lost PPA pricing, and expired equipment quotes? Those never make it onto the initial proforma. Most developers skip this: they treat interconnection as a post-siting checkbox instead of a siting constraint.

How far ahead should I look for grid data?

Three to five queue cycles, at minimum. A single snapshot tells you nothing—grid conditions shift faster than most solar irradiance models. One developer I know used a two-year-old cluster analysis, picked a site with 'available' capacity, and discovered the utility had re-dispatched that feeder for a data center load six months prior. The fix: subscribe to the local ISO's annual cluster study outputs, overlay planned transmission upgrades, and watch the trend in queued capacity, not the absolute number. If you see three consecutive cycles where available headroom drops below 15 MW, that site is already a trap—even if the parcel price per acre is half the county average.

Can tax incentives offset bad siting?

Rarely—and only in very specific windows. An ITC boost or a state-level production credit can mask a weak revenue projection for maybe two years. But incentives have expiration dates. The moment the tax equity flips or the legislative session ends, the site economics collapse back to their fundamentals: curtailment losses, upgrade delays, and O&M costs that climb faster than inflation.

'The worst mistake is treating a five-year incentive as a permanent fix for a thirty-year asset.'

— asset manager, after watching a 100 MW portfolio lose 14% IRR when a state PTC sunset

The catch is that some developers double down: they argue that a stepped-up incentive makes a weak site viable for the flip. That can work in a sale scenario if you find a buyer who values near-term cash over long-term risk. But the margin is razor-thin. One missed in-service deadline or a supply-chain hiccup that shifts the COD by a quarter—and the incentive window closes. Wrong order. That hurts.

What breaks first is usually the assumption that interconnection timelines will hold. I have seen projects where the developer used a 12-month queue projection, the site closed, and the utility study came back at 22 months. The ITC step-down had already triggered. The incentive math? Gone. The land was still cheap, though—a cold comfort when the project bank goes from funded to watch-listed in ninety days.