Your campus is permitted. Your tenant agreements are signed. Your equipment is on order. And your utility just told you the substation upgrade will take five to seven years.

This is no longer an edge case. It is the default experience for data center developers in capacity-constrained markets. The grid was not built for the load growth that AI and hyperscale computing are demanding, and the institutions that manage the grid are saying so publicly.

In November 2025, PJM’s Independent Market Monitor filed a formal complaint with FERC, arguing that large data centers should not be allowed to interconnect unless the grid can reliably serve them. The filing was blunt: adding massive loads without matching capacity forces the grid operator to allocate blackouts rather than ensure reliability. Their recommendation? “Bring your own generation” or face mandatory curtailment.

A month later, FERC commissioners called PJM’s failure to meet its reliability target in the 2026/2027 capacity auction “unacceptable.” The capacity market had already been strained by data center load growth—existing and forecast data center loads increased PJM’s capacity revenues by $16.6 billion across two auctions. Regulators, ratepayer advocates, and grid operators are now aligned on one point: the era of rubber-stamping data center interconnections is over.

Battery energy storage sits at the center of the solution. But not the way most people think.

Storage as Insurance, Not Arbitrage

The traditional pitch for co-located BESS at a data center goes something like this: install batteries, shave peak demand, capture some arbitrage revenue, maybe replace your diesel backup. It’s a C&I energy management story. A cost reduction play.

That pitch is obsolete.

The new pitch is this: a co-located BESS that smooths your ramp rates and provides dispatchable capacity transforms your data center from a grid problem into a grid solution. You stop being a load that threatens reliability. You start being an asset that strengthens it.

This is not theoretical. In October 2025, Aligned Data Centers and Calibrant Energy announced a 31 MW / 62 MWh BESS at Aligned’s campus in Hillsboro, Oregon—the first battery system in the U.S. purpose-built to accelerate a data center’s interconnection. The system allowed the facility to come online years earlier than traditional utility upgrades would have permitted. By converting the data center’s load from a “grid liability” into a “dynamic grid asset,” the project effectively bypassed the queue.

When you combine that precedent in the West with the regulatory pressure building in the East, the logic crystallizes: the killer app for data center BESS is no longer arbitrage. It’s access. The developers who build storage as an interconnection strategy—not as a cost optimization afterthought—will be the ones whose campuses come online on schedule.

Six Risks That Solar Playbooks Don’t Prepare You For

If you’re a data center developer evaluating BESS for the first time, or if you’ve hired a solar developer to “add storage,” here are the six categories of risk that will determine whether your project succeeds or stalls.

1. Interconnection Rejection

Grid operators are no longer processing large load interconnections on a first-come, first-served basis. PJM is actively developing rules that would require data centers to demonstrate that their load can be reliably served—or to bring their own generation. If your interconnection strategy is “file and wait,” you may be waiting indefinitely. A proactive BESS strategy gives grid operators a reason to say yes.

2. Fire Code Complexity Near Critical Infrastructure

A BESS next to servers worth hundreds of millions triggers different scrutiny than a standalone project in an empty field. Your Authority Having Jurisdiction will ask questions about thermal runaway propagation distances, explosion venting, and emergency shutdown logic that are specific to the proximity of critical computing infrastructure. The UL9540A test data that determines minimum separation distances must be validated against the local fire code interpretation of NFPA 855 before equipment is ordered. If your civil engineer starts designing the site before the Fire Protection Engineer establishes the fire safety design basis, you will redesign. We call this The Leeward Rule, and it is non-negotiable.

3. Duration Sizing Mismatch

Standard C&I BESS projects optimize around 4-6 hour demand windows for peak shaving. Your data center needs enough storage to bridge until backup generators stabilize or to ride through grid events while maintaining full computing load at 99.999% uptime. These are fundamentally different design problems. Undersizing means failing your SLA. Oversizing means stranded capital. The correct answer requires site-specific modeling against your actual load profile, local grid reliability data, and contractual uptime obligations—not a rule of thumb borrowed from a warehouse demand charge project.

4. BTM vs. FTM Regulatory Complexity

The Behind-the-Meter vs. Front-of-the-Meter decision isn’t binary, and getting it wrong locks in economics for twenty years. BTM simplifies permitting but forfeits wholesale market revenue. FTM enables a full value stack but introduces a separate interconnection process, different permitting requirements, and dispatch controls that must coordinate between facility backup and market signals. The Aligned project in Oregon actually ended up FTM because the utility said it could complete the studies just as fast—a counterintuitive outcome that only emerged through close collaboration. FERC’s December 2025 order directing PJM to create new co-location transmission services adds yet another configuration to evaluate. This is not a decision to make based on assumptions.

5. NERC CIP and Cybersecurity as a Performance Metric

FERC Order 901 directed NERC to develop comprehensive reliability standards for inverter-based resources, including BESS, with implementation rolling out through late 2026. For data center BESS—especially FTM configurations participating in wholesale markets—NERC CIP is not a compliance checkbox. Hyperscalers are intensely focused on grid-edge cybersecurity vulnerabilities. Your BESS EMS/SCADA architecture must support multi-factor authentication, local port locking, encrypted communications, and security logging as baseline features. These requirements must be specified in the OEM evaluation, not discovered during commissioning when it’s too late to change the control architecture.

6. Augmentation-Driven Cost Escalation

Batteries degrade. For a C&I peak shaving project, some capacity fade over time is acceptable. For a data center contractually obligated to maintain capacity for critical load, augmentation timing is driven by reliability requirements, not economics. This means augmentation happens earlier than a purely economic optimization would suggest, which changes how ownership is structured, who holds inventory risk, and what the long-term service agreement must cover. If the augmentation strategy is not modeled independently at the project’s outset—using site-specific weather data, the actual duty cycle, and the specific OEM’s degradation curves—the project’s twenty-year cost model will be wrong. Mixed-chemistry augmentation as technology evolves (NMC to LFP transitions, for example) adds EMS complexity that most basic platforms cannot handle.

Four Opportunities the Risks Create

The same forces creating risk are creating outsized opportunities for data center developers who move strategically.

BESS as Interconnection Insurance

We covered this above, but it bears repeating: co-located storage that provides dispatchable capacity and ramp rate smoothing is becoming a prerequisite for interconnection in constrained markets, not an optional add-on. The developers who deploy this strategy before the final FERC rulings drop will have signed the biggest supply agreements of 2026 while their competitors are still waiting for utility upgrades.

Front-of-Meter Revenue Stacking

If your BESS is sized beyond facility backup needs and configured for FTM participation, it becomes a revenue-generating asset in wholesale energy arbitrage, frequency regulation, and capacity markets. In PJM, where capacity revenues have surged specifically because of data center load growth, FTM BESS co-located at data center sites is exceptionally well-positioned. The sophisticated approach is a hybrid BTM/FTM configuration that captures both facility resilience and market revenue—but this requires navigating dual regulatory frameworks and coordinated dispatch controls.

Software and Controls as Competitive Moat

With battery hardware prices at record lows—roughly $105/kWh DDP for stationary storage—competitive advantage is shifting from procurement to the control layer. EMS platforms that support sub-second telemetry, real-time state-of-charge governance, multi-stream revenue optimization, and mixed-chemistry dispatch become the differentiator. For data center BESS specifically, the controls must coordinate between facility backup obligations and market dispatch. Projects that invest in sophisticated controls and can prove performance with data will command premium value, while projects running commodity EMS platforms will compete on price alone.

Utility Retirement Site Acquisition

As utilities retire coal plants and claim reuse of existing interconnection rights—Xcel Energy recently proposed doubling BESS capacity at its retiring Sherco coal plant from 300 MW to 600 MW, citing MISO rules requiring company ownership—a parallel opportunity emerges for data center developers. Co-locating near retiring thermal plants where transmission capacity is being freed up offers existing interconnection infrastructure, substations, and often willing host communities. The risk is that utilities will self-build and lock independent developers out. Speed-to-site and relationship capital become critical.

How We Work: The Owner’s Representative Model for Data Center BESS

Carina Energy is not an equipment vendor. We do not sell batteries. We are not an EPC contractor. We do not build projects.

We are your Owner’s Representative—the development management layer that coordinates the entire specialist team on your behalf. We manage the process so that professional liability stays with the insured specialists, not with you.

Our model is built on two principles:

Hyper-Local Execution: We hire the local civil and environmental firms who already have relationships with your Authority Having Jurisdiction. These are the engineers who know the planning commission members, who’ve permitted projects in the same county, who can get a meeting with the Fire Marshal without a six-week wait. Local knowledge is the most underpriced asset in development.

Expert Overlay: On top of the local team, we layer national BESS specialists—Fire Protection Engineers who interpret UL9540A test reports for a living, Owner’s Engineers who model degradation curves for breakfast, Interconnection Consultants who navigate queue reform across every ISO. These specialists don’t replace your local engineers. They inform them. The local civil engineer designs the grading plan; the FPE tells them what spacing to use. The local electrical engineer designs the single-line diagram; the BOE tells them what auxiliary loads to assume.

The coordination between these teams is managed through a Smartsheet-based project control system with a dedicated PMP tracking every dependency across the ~504-task development workflow. We own the Work Breakdown Structure and the critical path. We track earned value against consultant invoices to prevent overbilling. We inject risks identified during feasibility screening into the development management plan to ensure they’re mitigated, not forgotten.

The Phase Sequence

For data center BESS projects, the development process follows four phases:

Site Reality Check (48 hours): Before you sign an option or commit to a co-location site, we perform a desktop fatal flaw screen across civil, fire, and environmental constraints. If a setback eliminates buildable area, a wetland kills the layout, or a local moratorium blocks battery storage permitting, we find it before you spend capital.

Design Basis and OEM Selection: We issue a Technical RFI to qualified OEMs with a Fire Safety Annex requiring UL9540A test data and propagation distances. Our Fire Protection Engineer validates the test report against local code before the Design Basis Memo is issued. We never select a battery OEM without fire code validation first. This prevents the costly spacing redesigns that have delayed campuses across the country.

Design, Permitting, and Interconnection: The local EOR handles grading, stormwater, and the permit application. The FPE produces the Hazard Mitigation Analysis and coordinates with the Fire Marshal. The Interconnection Consultant manages queue position, utility studies, and NERC CIP classification. The BESS Owner’s Engineer independently models degradation and defines the augmentation strategy. Our PMP coordinates all of these workstreams through Smartsheet, ensuring no handoff is missed and no dependency is broken.

Procurement and NTP: Once the design is permitted, we issue a Commercial RFP for final pricing. This is the RFI/RFP split: we do not lock in lithium pricing during the design phase. The FPE performs a forensic review of the un-redacted UL9540A report for the winning bidder. The Owner’s Engineer reviews the Battery Supply Agreement’s technical exhibits. You get competition at the right time, with technical validation at every gate.

Why Not Just Use the OEM’s Team? Or Build Internal?

Two objections we hear frequently. Both are reasonable. Neither survives scrutiny.

The OEM objection: “Our battery supplier offers turnkey development services.” They do—and they’re evaluating their own product. An OEM-led development process will never recommend a competitor’s technology, even if it’s a better fit for your site’s fire code constraints. They will not perform an adversarial review of their own UL9540A test data. And when their spacing requirements don’t fit your parcel, they’ll suggest you buy a bigger parcel rather than evaluate a different battery. An Owner’s Representative works for you, not for the equipment vendor.

The internal team objection: “We’ll hire a BESS development manager.” You should—eventually. But that person still needs an FPE, a BOE, an interconnection consultant, and a local EOR. They still need to coordinate those disciplines across the 504-task workflow. And they need to know which OEM claims to challenge, which consultant estimates to validate, and which AHJ interpretations to push back on. One person, no matter how talented, cannot carry all of that institutional knowledge. The Owner’s Representative model gives you the team and the operating system while your internal capability matures.

The Takeaway

Data center BESS is not an oversized C&I project. It is not “solar plus batteries.” It is its own category—closer to transmission-scale complexity with C&I-scale footprints—and it requires a development process designed for its specific risks.

The regulatory environment is accelerating. FERC is actively developing rules for co-located loads. Grid operators are tightening interconnection standards. Capacity markets are repricing around data center demand. The window to deploy storage as a proactive interconnection strategy—rather than a reactive compliance obligation—is open now, and it will not stay open indefinitely.

The developers who move first will move fastest. The ones who wait will wait longest.

Carina Energy provides BESS Development-as-a-Service for data center developers, including Owner’s Representative services, fatal flaw screening, and Smartsheet-based project controls. Contact us to discuss your project, or request a 48-hour Site Reality Check before committing capital to your next site.