Big decisions deserve careful questions.

The proposed data center near North Polzin Road and West Fenric Road presents a consequential land-use decision for the Town of Janesville. Data centers provide essential digital infrastructure, but the merits of the industry do not establish that this site, this project, or its proposed terms are appropriate for the community.

This manual offers a neutral, evidence-based framework for evaluating the proposal before irreversible decisions are made. Priority questions concern compatibility with neighboring homes and farmland; the availability of industrial or previously developed alternatives; electrical demand and infrastructure costs; nighttime noise; water use and groundwater protection; and the net, long-term economic benefit to residents.

A further concern is financial execution risk. Announced capital expenditure is not the same as secured financing or completed construction. Industry-wide uncertainty about proposed data-center capacity warrants project-specific scrutiny, not an assumption that this developer will fail. Officials should establish who is financially responsible, which financing and customer commitments are verifiable, and who bears costs if the project is delayed, scaled back, transferred, or abandoned.

Before approval or public financial commitments, citizens should seek:

• Independent technical assessments and realistic comparisons with alternative sites.
• Disclosure of projected full-build electrical and water demand, noise impacts, and infrastructure requirements.
• A 20-year fiscal assessment covering full-build, partial-build, delayed, and abandoned scenarios.
• Legally appropriate, enforceable safeguards for taxpayers, utility customers, nearby residents, and the land itself.

The central test is straightforward: Is this the right location, is the project credible and beneficial, and are the community’s interests protected if projections prove wrong? The objective is neither automatic approval nor automatic rejection, but a decision supported by evidence and accountable commitments.

This manual is written specifically for citizens interested in the proposed data center near North Polzin Road and West Fenric Road in the Town of Janesville. It is not intended to support or oppose the project. Its purpose is to identify the issues that deserve the greatest attention before local officials make a long-term land-use decision. The emphasis is on practical questions, evidence, and the specific characteristics of this proposal rather than on generic discussions of data centers.

A well-supported decision would be based on independent technical studies, transparent disclosure of operating assumptions, public review of environmental information, enforceable permit conditions, and ongoing monitoring. Whether the final decision is approval or denial, confidence in that decision increases when it is supported by evidence rather than assumptions.

The proposed Town of Janesville data center should be evaluated as a land-use decision rather than as a referendum on technology. The central question is whether the proposed location, operating characteristics, and community protections represent the best long-term outcome for Janesville. That determination requires complete information, careful analysis, and measurable commitments.

At the time of writing, discussion has centered on a proposed data center on farmland in the Town of Janesville. Public reporting indicates that the proposal is in an early stage and that many important details remain unavailable. Unlike many summaries of data centers, this manual begins with what is known and what remains unknown because the unanswered questions are often as important as the known facts.

• Known:
• The proposed location is on agricultural land near existing residences.
• The site appears attractive because of nearby electrical transmission infrastructure.
• Separate data-center discussions have also involved redevelopment of the former GM/JATCO industrial property.
• Unknown:
• Ultimate electrical demand.
• Cooling technology.
• Average and peak water demand.
• Permanent employment.
• Detailed acoustic modeling.
• Stormwater and groundwater monitoring plans.

The principal planning question is not whether data centers are beneficial. Modern society depends on them. The more important question is whether this particular location is the best available location. The presence of existing industrial land elsewhere in Janesville naturally raises the question of whether redevelopment could reduce conflicts between industrial activity and residential neighborhoods.

Myth: Cooling water is contaminated by contact with CPUs and GPUs.

Reality: Modern cooling systems isolate coolant from electronic components. Routine contaminants originate primarily from dissolved minerals, treatment chemicals, corrosion products, and occasional spills.

Myth: Trees solve the noise problem.

Reality: Trees provide visual screening but limited reduction of low-frequency industrial noise. Distance, berms, acoustic treatments, and equipment selection are more effective.

Myth: A billion-dollar data center creates thousands of permanent jobs.

Reality: Construction employment and permanent operating employment are very different. Permanent staffing is typically much smaller than many people expect.

Myth: Every concern has equal importance.

Reality: For this proposal, electrical demand, land use, noise, water demand, and groundwater protection appear more significant than concerns about electronic contamination.

Electrical Infrastructure

The strongest long-term concern is likely to be electrical demand. Citizens should seek clear answers regarding peak demand, full build-out, substations, transmission upgrades, and responsibility for infrastructure costs.

Continuous Noise

Cooling equipment operates continuously. Nearby residents are affected by nighttime and low-frequency sound rather than short bursts of loud noise. Independent acoustical studies and enforceable limits are more valuable than general assurances.

Water Demand

Southern Wisconsin is not the desert Southwest, but water demand should still be disclosed. The important questions are how much water will be required, where it will come from, and how demand changes during hot weather.

Aquifer Protection

Routine contamination from CPUs and GPUs is not expected to be a meaningful issue. Greater attention should be directed toward spill prevention, cooling-water treatment chemicals, groundwater withdrawals, stormwater management, and baseline monitoring of nearby wells.

Land Use

The proposed location should be evaluated against realistic alternatives, including previously developed industrial land, not merely against the option of doing nothing.

Economics

Construction employment should not be confused with permanent operating employment. Long-term fiscal benefits should be evaluated after considering tax incentives and infrastructure costs.

Why financial feasibility matters

Announced investment is not the same as committed investment. Industry forecasts suggesting that substantial announced data-center capacity may not be completed are a reason to examine this project's specific financing and execution risks—not evidence that this particular project will fail. Before the community relies on projected investment, jobs, and tax revenue, it should establish which commitments are binding and which remain contingent.

Developer and customer commitments

Identify the legal development entity, ultimate ownership, intended operator, and party responsible for infrastructure obligations. Request credible evidence of project financing, construction milestones, and customer demand. Confidential agreements need not be publicly disclosed in full; an independent reviewer could verify material commitments and report findings without exposing commercially sensitive details.

Infrastructure and stranded-cost exposure

Determine which transmission, substation, road, water, and wastewater improvements are required, who authorizes them, and who pays under full-build, partial-build, delayed, and cancelled scenarios. Ask utilities and regulators about minimum-billing provisions, collateral, termination payments, and protection of existing ratepayers. Do not assume local zoning authorities control utility tariffs.

Risk-adjusted economic benefits

Expand the proposed 20-year fiscal analysis to compare full completion, partial completion, extended delay, and abandonment. Distinguish construction employment from permanent jobs, gross investment from taxable local value, and projected tax receipts from net public benefits after incentives and public costs.

Phasing and enforceable protections

Where authorized under Wisconsin law, consider phased approvals tied to demonstrable milestones, performance security for publicly funded work, site restoration requirements, and enforceable responsibility for unfinished infrastructure. Determine which conditions can lawfully be imposed through zoning, development agreements, utility proceedings, or incentive contracts.

A balanced conclusion

Financial uncertainty does not justify automatic rejection. It does justify declining to make irreversible public commitments based solely on nonbinding projections. The same evidentiary standard should apply to developers that favor the project and citizens who question it.

Experience elsewhere provides useful tests for Janesville’s assumptions. These examples are not evidence that the proposed Janesville project will encounter the same problems; they illustrate why independently verifiable commitments and enforceable safeguards matter.

Mount Pleasant, Wisconsin — Construction pauses and employment claims

Microsoft reported approximately 550 full-time people on-site supporting its first Mount Pleasant facility in June 2026, while its own general guidance describes roughly 50 employees and vendors per typical data center building. The 550 figure does not isolate direct employees from contractors or establish a representative staffing level for Janesville. During startup, nearby residents also reported a tonal hum from cooling fans. Microsoft acknowledged the issue, implemented adjustments, and committed to further mitigation. The first facility became operational; subsequent campus development continued. [1–3]

Grant County, Wisconsin — A proposal that did not materialize

Wisconsin Watch documented a proposed 500-acre, up-to-$2-billion data center project that appeared to collapse after developers stopped responding to local economic-development officials. The reporting does not establish that the county incurred significant stranded infrastructure expenses. It does show why projected investment should not be treated as a binding commitment. [4]

Port Washington, Wisconsin — Infrastructure plans under scrutiny

On September 10, 2026, Wisconsin’s Public Service Commission revoked the completeness determination for a transmission-interconnection application serving the Port Washington development. Regulators required a new application after extensive changes hindered meaningful review. Reporting placed estimated infrastructure costs at $1.3–$1.7 billion. This was a regulatory restart, not proof of abandoned construction or realized ratepayer losses. [5–6]

Fayette County, Georgia — Operational water oversight

E&E News by POLITICO reported that nearly 30 million gallons of water consumed at a QTS campus had not been billed, following complaints about low pressure. The utility assessed $147,474 in retroactive charges; QTS reported paying them, and officials said metering had been corrected. This is an observed operational oversight issue, not evidence of an enduring unpaid public loss. [7]

Washington state — Incentives and verification

A Seattle Times–ProPublica investigation found that Washington’s data center tax exemption was expanded while hiring requirements were relaxed and facility-level employment outcomes were not publicly available. The reporting identified serious verification gaps, not proof that recipients created no jobs. [8]

Employment promises: a question for every case

Microsoft’s general guidance describes approximately 50 employees and vendors per operating data center building, but staffing varies by design, size, and campus. Janesville should require separately reported peak and cumulative construction employment, steady-state operating FTEs, contractors, local hiring, and annual verification. Compare any incentive cost per verified permanent job with a full 20-year fiscal analysis. [1–2, 8]

Documented sources and references

Numbered references correspond to the community case studies above. Official sources and established investigative or public-service reporting are prioritized. Sources describe outcomes as of October 23, 2026; future developments may change project status.

[1] Microsoft, “Frequently asked questions about our datacenters” (typical staffing). Open source

[2] Microsoft, “Microsoft completes construction on first datacenter facility in Mount Pleasant, Wisconsin,” June 23, 2026. Open source

[3] Microsoft, “Update on our investigation of humming sound at our Mount Pleasant datacenter,” updates April–July 2026. Open source

[4] Wisconsin Watch, “A $2 billion proposal, then silence: How a Driftless Area data center deal fell apart,” June 22, 2026. Open source

[5] Public Service Commission of Wisconsin, Ozaukee County Distribution Interconnection Project, docket 137-CE-221, September 10, 2026 order. Open source

[6] Wisconsin Public Radio, “PSC deals major blow to transmission project supporting Port Washington data center,” 2026. Open source

[7] E&E News by POLITICO, “A data center drained 30M gallons of water unnoticed — until residents complained about low water pressure,” 2026. Open source

[8] ProPublica / The Seattle Times, “A Tax Break for Washington Data Centers Promised Jobs. Is It Paying Off?” 2024. Open source

• Why was this site selected over existing industrial or brownfield property?
• What is the ultimate build-out of the campus?
• What is the maximum electrical demand?
• Who funds utility upgrades?
• What are the average and peak water demands?
• What cooling technology will be used?
• What are the predicted nighttime noise levels at the nearest homes?
• Will baseline private-well testing be offered?
• Will groundwater monitoring wells be installed?
• What chemicals will be stored onsite?
• What are the permanent staffing projections?
• What tax incentives are requested?
• What is the twenty-year net fiscal benefit?
• What corrective actions are required if predictions prove inaccurate?

Additional financial due-diligence questions

• Which entity is contractually responsible for completing the project and paying for dedicated infrastructure?
• What financing and customer commitments have been independently verified, and what remains conditional?
• What is the project schedule for each phase, and what triggers further construction?
• Who bears costs if electrical demand or construction falls short of forecasts?
• What security, minimum payments, and termination provisions protect utility customers and taxpayers?
• How do 20-year net fiscal outcomes change under full, partial, delayed, and cancelled development?
• What obligations survive a sale, restructuring, or insolvency of the development entity?