What makes a data center a good neighbor?

This is an original audio briefing from Good Neighbor Data. It is narrated with an AI-generated voice from a script prepared for our Responsible Data Center Resource Center.

Data centers support cloud services, communications, health care, business, government, entertainment, and much of everyday digital life. They can also bring major investment, construction work, permanent technical roles, and new tax revenue.

But those benefits do not answer the local question.

A community still needs to know how a proposed project will affect its power system, water supply, land, roads, public services, nearby homes, and long-term obligations.

That question is becoming more urgent. Good Neighbor Data's public map documented one hundred forty-two active or extended local restrictions on data center development, plus twenty-two proposed or pending actions, as of July twenty twenty-six. These actions differ in scope and legal form. Together, they show that communities want clearer evidence before commitments become fixed.

So, what makes a data center a good neighbor?

The short answer is local fit, transparent tradeoffs, and commitments that can be measured.

No single technology, efficiency target, or sustainability claim can establish that on its own. A responsible review begins with the maximum planned project, including every phase and the supporting infrastructure. It separates estimates from confirmed commitments. And it asks who is responsible for each outcome over time.

Start with power.

Annual electricity use is not the same as peak demand. A facility may use a large amount of energy over a year, while the grid must still be designed for the highest demand that occurs at a particular hour. Local generation, transmission, substations, other planned loads, and the timing of new infrastructure all shape the answer.

A credible power plan explains average demand, peak demand, the hourly profile, the phase schedule, and the uncertainty in the forecast. It also identifies required upgrades, who pays for them, what happens if the project is delayed or downsized, and whether the facility can reduce demand during constrained periods.

The right question is not simply, “Does this use a lot of electricity?” It is, “Can this system serve the project reliably, with costs and risks assigned transparently?”

Next is water.

There is no universal number for data center water use. Demand changes with facility size, server efficiency, utilization, cooling technology, climate, operating conditions, water source, and the electricity supplying the site.

Water and energy decisions are connected. Dry cooling can reduce routine on-site water consumption, but may require more fan energy or perform less efficiently in hot conditions. Evaporative cooling can reduce cooling energy, but consumes water through evaporation. Direct liquid cooling can move heat efficiently near computing equipment, yet the complete water impact still depends on how the captured heat is rejected.

Berkeley Lab found that workload-level water use can vary by more than ten thousand times across combinations of server efficiency, grid water factors, utilization, cooling, infrastructure efficiency, climate, and other conditions.

That is why communities should request average and peak demand, the water source, wastewater effects, drought operating plans, direct and indirect use, full-buildout scenarios, and meaningful metering after opening.

Then consider sound, land, and nearby uses.

Data center sound can come from cooling equipment, fans, transformers, pumps, generators, vehicles, construction, and maintenance. Overall loudness matters, but so do low-frequency sound, tonality, time of day, equipment combinations, terrain, weather, and the quietness of the existing environment.

A useful acoustic review follows a source, path, and receiver framework. Choose quieter equipment and controls at the source. Use orientation, distance, buildings, barriers, enclosures, and vibration isolation along the path. Then evaluate the actual receivers: homes, schools, parks, places of worship, and other sensitive uses.

The same local discipline applies to land. The footprint includes more than the main building. Substations, transmission, water and wastewater facilities, roads, grading, stormwater systems, security, lighting, buffers, and landscape maintenance can shape the place for decades.

A good-neighbor project studies these systems before the layout and equipment choices are fixed, then verifies the completed design under defined operating conditions.

Community value also needs evidence.

Data centers can expand the property-tax base and support substantial construction work. But permanent staffing is usually much smaller than construction employment. Net fiscal benefit depends on valuation, depreciation, tax rates, incentives, infrastructure obligations, and public-service costs.

Communities should ask for an independent fiscal analysis, not only a headline investment number. Job estimates should separate construction from operations, direct hires from contractors, and temporary work from permanent roles. Infrastructure and community commitments should identify the responsible party, delivery schedule, maintenance obligation, and reporting method.

Finally, accountability turns a promise into a good-neighbor commitment.

Every material commitment should answer five questions.

What exactly will be measured?

Where and when will it be measured?

Who is responsible?

How will the result be reported?

And what happens when performance differs from the commitment?

Those answers can inform permits, development agreements, utility agreements, commissioning plans, public reports, and corrective procedures, subject to local authority and qualified technical and legal review.

The obligations should also survive ownership changes. A complete plan addresses assignment, repowering, reuse, closure, equipment removal, remediation, and any financial assurance needed to protect the community.

A data center earns trust when the chain is clear: a local concern connects to a project decision, the decision becomes a measurable commitment, and the operating result is verified.

That is the Good Neighbor Data standard. Accountability over slogans.

For practical evidence requests covering power, water, sound, land, infrastructure, fiscal effects, emergency response, and long-term reporting, visit the Responsible Data Center Resource Center and use the planning checklist at good neighbor data dot com.

This briefing is educational and is not legal, engineering, utility, environmental, fiscal, or planning advice. Project decisions should be reviewed for the specific site, jurisdiction, utility territory, and operating plan.