Pataskala, Ohio Killed Its Data Center Three Times This Year. Only the Last Vote Can't Be Undone.
Pataskala's council already killed Aligned's 200-megawatt proposal in August. On Tuesday, 73.6% of voters wrote the ban into the city charter anyway -- one of 41 data center measures on ballots in 10 states, nearly all of which broke the same way.
In Pataskala, Ohio, a data center died twice before most of the country noticed it had died at all. In June, the city's planning commission voted 5-2 to recommend denying Aligned Data Centers' site plan for an 89-acre, 200-megawatt campus, after hours of public questions the commission couldn't get answered to its satisfaction. In August, the city council made it official, voting unanimously to reject the project outright. On Tuesday, the voters of Pataskala showed up anyway, and 73.62% of them -- 5,418 to 1,941, with all precincts reporting -- amended the city charter to ban any data center over 25 megawatts from ever building there again.
That's not a protest vote against a live threat. By November, there was no project left to stop. It's something closer to an insurance policy: a charter amendment outlasts a city council, survives an election cycle, and can't be quietly reversed by five votes the way an ordinary zoning denial can. Pataskala's voters weren't relitigating August. They were making sure no future council gets the chance to say yes to the next one.
Pataskala and Granville sit in Licking County, just east of Columbus, which is not some rural holdout that's never seen a data center up close. It's the next county over from New Albany, where Meta's 1,000-megawatt Prometheus campus is under construction, Amazon has bought 439 acres along Beech and Miller roads, Google has spent $63 million on its own site, and Intel is building a $28 billion chip plant. Voters here weren't rejecting data centers as an abstraction. They were drawing a line a county away from the real thing, already running at a scale most of the country has never seen.
The project that lost three times, in the same town
The fight started over specifics, not abstractions. Aligned wanted 89 acres for a facility residents worried would strain Licking County's water supply and electric rates long before it created many local jobs. "We need to see exactly what we're voting on," council member Mary Hite said before the vote, a complaint about the project's vagueness that ran through most of the public record. Resident Leatrice Guttentag put it more bluntly: "They just don't belong near residents." Another resident, Grace Franklin, pressed on whether there would be enough water to go around and whether electric rates would rise to pay for infrastructure the rest of the city never asked for. Council President Brandon Galik said he voted no because too many of those questions never got answered, which is a council member describing a failed diligence process in one sentence, whether he meant it that way or not.
Eleven miles away, Granville had no project pending at all, and its voters passed an identical charter ban anyway, 82.45% to 17.55%. Tracee Laing, who organized the petition drive there, called her town's measure "largely symbolic" and said it would matter more in Pataskala, "where they are facing data center proposals." Both towns used ballot language supplied by Conserve Ohio, a statewide group that has clearly decided the charter amendment, not the one-off zoning fight, is the durable way to win this argument for good.
This didn't start in Ohio. Back in June, Monterey Park, California became the first city in the country where voters, not a council, enacted a permanent data center ban at the ballot box, approving Measure NDC by nearly 86%. Councilmember Jose Sanchez argued at the time that a public vote carries more legal and political weight than an ordinance five council members could pass or reverse: "Being able to go to court and say the residents of Monterey Park voted to ban data centers is a much better gauge of where our residents are." Khara Boender of the Data Center Coalition, the industry's own trade group, warned that the ban would cost the area jobs, investment, and tax revenue. Pataskala and Granville just ran the same play five months later, on a much larger, more organized scale.
Ohio wasn't the outlier. It was the biggest sample size.
Ohio put 21 separate data center measures in front of voters this week, more than any other state, spread across 18 cities, villages, and townships in and around Columbus. Nearly all of them passed. A statewide poll taken before the election found 55% of Ohioans opposed new data center construction and only 22% in favor, and on Tuesday the vote mostly caught up with that sentiment. A parallel push to put a constitutional ban in front of the entire state fell short of the 413,487 signatures it needed by July; organizers are already aiming for November 2027.
Ohio was the biggest piece of something much larger. Ballotpedia counted 41 local data center ballot measures across ten states on Tuesday -- California, Florida, Illinois, Kansas, Maine, Nevada, New Jersey, North Dakota, Ohio, and Wisconsin -- ranging from outright bans like Pataskala's to new disclosure and water-use rules short of a ban. Nearly all of them tightened the rules or stopped the facilities outright. The lone partial exception, in Harwood, North Dakota, wasn't actually a vote for a data center. Voters there rejected a measure that would have added disclosure and public-review requirements to industrial projects, including data centers, which is a vote against more paperwork, not a vote inviting anyone in. By any honest count, this was close to a sweep.
It's worth saying what this sweep wasn't. It wasn't a single industry group outspending a scrappy local campaign and losing anyway. Keith Durkin of Ohio Northern University, who watched the Licking County races up close, said he'd "never seen any pro-data center advertising" in the run-up to the vote. Boender and the Data Center Coalition made their case in national press statements, arguing these bans cost residents real job opportunities and long-term investment, and that argument isn't wrong as a general matter. But it showed up after the fact, in trade coverage, not on yard signs in Pataskala or mailers in Granville. A case made only in the aftermath doesn't move a single vote.
What this actually means if you're siting one
I think the mistake is treating all of this as a communications problem, the kind you fix with a better-funded local campaign next time. Pataskala's planning commission flagged real trouble in June. The city council confirmed it in August. The charter amendment in November was the third and most durable verdict on the same set of facts, not a surprise to anyone who'd been reading the public meeting record. If you were doing real diligence on that site, the signal to walk away showed up months before a single vote was certified.
Political risk has always belonged in the same evidence ladder as an interconnection study, a water right, or a zoning entitlement, graded by how current and how documented it actually is, and this week is the clearest argument yet for treating it that way instead of as background noise. A signed utility tariff doesn't survive a charter amendment that bans the facility outright, and neither does a land option. The towns writing these bans into their charters, rather than just voting down one proposal, are telling developers something specific: don't assume the next council, or the one after that, will see it differently.
For anyone actually underwriting a site right now, the ballot box isn't a wildcard anymore. It's a diligence item, and a checkable one, months before Election Day, if you bother to read the planning commission minutes the way Pataskala's own commission already had back in June. Barely twenty miles from there, New Albany's grid is already being rebuilt to carry gigawatts of the exact load Pataskala just voted to keep out. Whether that gap closes with a well-sited project or stays a line on a map now runs through the same county, under two very different verdicts.
Evaluating a specific parcel for a 50–300 MW load? GridPermit's Preliminary Site Kill Test grades the power, zoning, water, and commercial-readiness evidence before you buy the land.
Start a Qualification ReviewSee state-by-state coverage · Read the methodology · See a sample report