Last quarter, an institutional owner-operator approved a $2.4 million chiller replacement with a 3.7-year payback. The capital committee spent under twenty minutes on it. The same committee then debated a $380,000 building analytics platform for three hours—and tabled it. This isn’t an outlier. Across the portfolios I analyze, the pattern repeats: physical plant gets funded with mechanical precision; digital intelligence gets interrogated with the suspicion usually reserved for speculative R&D.
The conventional explanation blames the technology. The platform is unproven. The ROI is fuzzy. The vendor couldn’t show a clean IRR. But that explanation collapses under scrutiny. The chiller vendor’s energy-savings projections relied on nameplate efficiency gains, not measured performance. The analytics vendor brought twelve months of fault-detection logs from a comparable portfolio showing $410,000 in avoided energy and repair costs—actual data, not engineering estimates. The difference in outcome wasn’t about the evidence. It was about the framework the committee used to evaluate it.
This is the core problem I want to unpack. The gap between funding a chiller and funding a building analytics platform is not technological. It is a structural failure in how organizations underwrite operational intelligence as an asset class. The same CFO who treats a cooling tower as depreciable hard property with a known useful life will treat a fault-detection dashboard as a software subscription—something that belongs in an IT budget, not a capital plan. The accounting treatment, the approval pathway, and the mental model are all wrong for what the technology actually does.
The Capital Allocation Architecture Problem
Most large organizations operate with split budgets: capital expenditure for assets with useful lives beyond one year, and operating expenditure for everything else. A chiller fits neatly into capex. It gets capitalized on the balance sheet, depreciated over fifteen to twenty years, and evaluated against a hurdle rate for return on invested capital. The decision framework is mature, standardized, and legible to the board.
A building analytics platform doesn’t fit. The hardware—sensors, gateways, edge devices—might qualify as capitalizable. But the software, the configuration, the ongoing tuning, the data integration layers? Those get classified as opex, as a maintenance contract, or worse, as “IT spend.” The capital committee looks at the proposal and sees an expense that will hit the P&L immediately, with no depreciable asset to show for it. The evaluation framework collapses. There is no line item for “decision quality improvement” or “operational risk reduction.” There is no standard depreciation schedule for a fault-detection algorithm.
This isn’t a minor accounting nuance. It determines which projects survive the approval gauntlet and which die in the finance review. When a facilities VP brings a chiller proposal to the CFO, she speaks the CFO’s language: capital outlay, useful life, maintenance savings, energy reduction, simple payback. When the same VP brings a building analytics proposal, the VP often defaults to the vendor’s language: dashboards, alerts, machine learning, fault detection. The CFO hears “software” and mentally moves it to a different approval track—one with lower spending authority, annual renewal scrutiny, and zero strategic weight.
The solution starts with reclassifying the investment. Building analytics is not software. It is a decision-support asset that changes how the organization allocates maintenance capital, schedules equipment replacement, and manages energy procurement. The useful life of the insight it generates extends well beyond a single budget cycle. Organizations that have successfully funded these platforms rewrote the narrative: they treated the analytics investment as a capital project with an identifiable return stream tied to avoided costs, deferred capital, and reduced energy spend. They built a pro forma that looked like a chiller proposal—not an IT purchase order.
Why Payback Period Is the Wrong Metric for Intelligence
The second structural barrier is the dominance of simple payback as a decision filter. For a chiller, payback works reasonably well. You know the installed cost, the efficiency gain over the existing unit, and the expected energy savings. The calculation is clean, if optimistic. But payback assumes the benefit stream is singular and predictable: lower energy bills, year after year.
A building analytics platform generates a portfolio of benefit streams, many of which are non-linear and contingent on organizational action. Fault detection identifies a stuck damper that was wasting $18,000 annually in reheat energy. That’s a direct energy saving. But the same platform also identifies a chiller sequencing error that, left uncorrected, would have shortened compressor life by four years—avoided capital replacement cost of $140,000. It flags a simultaneous heating and cooling condition that, once fixed, reduces maintenance dispatches by thirty per year—avoided opex. It surfaces occupancy patterns that justify downsizing a planned HVAC upgrade—avoided future capex.
These benefit streams don’t arrive on a predictable schedule. They require a human to act on the insight. The ROI is real but path-dependent: it depends on whether the facilities team has the capacity and authority to respond to the alerts. A payback model can’t capture this optionality. It treats the analytics platform as a cost-reduction tool rather than an option on better capital allocation across the entire portfolio.
The better framework is net present value over the hold period, with explicit scenario modeling. What is the probability-weighted value of avoiding one major equipment failure over five years? What is the value of deferring a capital replacement by eighteen months because the analytics platform proved the existing asset had more remaining useful life? These questions require probabilistic thinking, not point estimates. They also require the finance team to accept that some benefit streams—reduced tenant complaints, faster lease-up, lower insurance premiums due to better risk management—are real even if they resist precise quantification.
The Conflation of Software With Maintenance Contracts
A third distortion comes from procurement’s tendency to categorize building analytics as a maintenance service. Many platform vendors bundle ongoing tuning, algorithm updates, and support into an annual fee. To the procurement team, this looks like a service contract—something to be benchmarked against other FM service contracts, negotiated down, and renewed annually. The strategic value of the data asset gets buried in a category designed for mowing lawns and changing filters.
This categorization has real consequences. Service contracts rarely survive cost-cutting exercises intact. When the CFO demands a 10% opex reduction, the facilities director looks at the analytics platform fee and sees a line item with no immediate operational consequence if cut—unlike, say, the janitorial contract. The platform gets cancelled, the sensors go dark, and three years of operational data that could have informed the next capital plan evaporates. The organization saves $380,000 in opex and loses the ability to avoid $2 million in misallocated capex over the next five years. The accounting framework made the trade invisible.
Organizations that sustain these investments treat the analytics platform as a capital asset with a multi-year commitment, not an annual service. They fund it from the capital budget, amortize it over a defined period, and protect it from opex cuts. They also write the contract differently: multi-year terms with performance guarantees, data ownership clauses that survive termination, and handover protocols that ensure the organization retains the data and the insights even if the vendor relationship ends.
The Narrative Capability Gap
Underneath all of these structural barriers is a capability problem. The facilities and engineering teams that understand building performance lack the financial vocabulary to make their case. The finance teams that control capital allocation lack the operational context to see why a dashboard matters. The result is a proposal that gets lost in translation.
I’ve seen this gap manifest in board presentations where the facilities director shows a slide of fault-detection alerts—impressive in volume, meaningless in financial impact. The CFO, who thinks in EBITDA and return on assets, sees noise. What the CFO needed was a single slide showing: “This platform will reduce our portfolio energy spend by 8-12%, defer $4.2 million in planned HVAC replacements by surfacing remaining useful life, and prevent an estimated $800,000 in emergency repair costs over five years. The NPV at our 10% hurdle rate is $1.7 million. We recommend capitalizing the platform cost and amortizing it over seven years, consistent with the expected useful life of the insight generation capability.”
That slide did not require better technology. It required someone on the team to translate operational intelligence into the language of investment decision-making. This is not a skill most facilities organizations possess. But it can be developed—through joint workshops with finance, through hiring analysts who bridge both worlds, and through a deliberate effort to build the business case narrative before the vendor proposal arrives.
In some ways, the challenge resembles what writers face when adopting new tools: the technology may be powerful, but unless the output reflects genuine human judgment and context, it fails to persuade. The Authors Guild, in its AI Best Practices for Authors, emphasizes that the writer’s original voice and thinking remain the source of value—AI outputs are generic mashups unless shaped by human discernment. The parallel for corporate real estate is exact: a building analytics platform generates alerts; it takes a skilled operator to interpret them, prioritize them, and translate them into financial decisions that a board can evaluate. The technology is an input, not the output. The narrative—the human translation—is where the investment case lives or dies.
A Due Diligence Framework for Building Intelligence Investments
Given these structural barriers, how should a corporate real estate or strategy team evaluate a building analytics platform? I propose a framework that treats the investment as a strategic commitment, not an IT purchase. The framework has four dimensions: asset classification, benefit quantification, contract structure, and organizational readiness.
Asset classification. Before evaluating ROI, decide how the investment will be treated on the books. Will you capitalize the hardware and expense the software? Will you treat the entire platform as a capital project with a defined amortization period? Will you create a new asset category for operational intelligence systems? The accounting treatment determines which approval pathway the proposal will travel—and whether it will survive the journey. Work with your controller’s office early, not after the vendor proposal lands.
Benefit quantification. Move beyond simple payback. Build a five-year NPV model with at least three benefit streams: direct energy savings from fault correction, avoided capital replacement from extended equipment life, and avoided operating costs from reduced emergency maintenance. For each stream, use ranges rather than point estimates. Model a base case, a conservative case (where only 60% of faults get corrected), and an optimistic case. Show the probability-weighted expected value. This is the format your CFO already uses for other investment decisions.
Contract structure. Negotiate the contract as a performance partnership, not a software subscription. Include minimum performance guarantees tied to measurable outcomes: fault detection rate, false positive percentage, energy savings attribution. Require data portability and a handover protocol if the vendor relationship ends. Structure payments to align with value delivery: a portion of the fee contingent on verified savings. This shifts the vendor’s incentive from selling licenses to delivering results.
Organizational readiness. The best platform will fail if the facilities team lacks the capacity to act on the insights. Before funding the technology, assess whether your team has the skills, the bandwidth, and the authority to respond to fault alerts. If not, budget for training, process redesign, or a managed service layer. The investment case should include not just the platform cost but the cost of building the human capability to use it.
This framework is not a theoretical exercise. I have seen it applied in a portfolio of twenty-two office buildings where the analytics investment was initially rejected as “IT spend.” The facilities director rebuilt the proposal using the framework, secured capital committee approval, and delivered a 23% IRR over three years—measured against actual avoided costs, not vendor projections. The difference was not the technology. It was the case.
Why This Matters Beyond the Single Investment
The chiller-versus-analytics problem is a microcosm of a larger issue in corporate real estate. As buildings become more data-intensive, the gap between what the technology can do and what the organization can evaluate will widen. Digital twins, automated fault detection, predictive maintenance, integrated workplace management systems—these represent a category of investment that does not fit the traditional capex/opex binary. They generate returns that are probabilistic, path-dependent, and contingent on organizational behavior. If we don’t fix the evaluation framework now, we will systematically underinvest in the very capabilities that could make our portfolios more resilient, more efficient, and more valuable over the hold period.
The fix is not to demand better ROI models from vendors. It is to build the internal capability to translate operational intelligence into the language of capital allocation. This means putting analysts on the real estate team who can build NPV models, not just energy models. It means training facilities directors to present to finance committees, not just engineering conferences. It means demanding that procurement treat building analytics contracts as strategic partnerships, not commodity services. And it means recognizing that the value of a building analytics platform is not in the software—it is in the decisions the software enables.
One final observation. In my work analyzing corporate portfolios, I’ve noticed that the organizations most successful at funding these investments share a specific trait: they have someone—often a director of portfolio strategy or a VP of real estate—who has learned to write the narrative. They can take a fault-detection log and turn it into a slide that shows avoided capital, deferred opex, and reduced risk exposure in terms a board member recognizes. This skill is rare, but it is learnable. The Purdue Online Writing Lab’s resources on Creative Writing underscore a principle that applies here: effective writing is about shaping raw material into a coherent, persuasive form. The raw material in this case is sensor data and maintenance logs. The persuasive form is a capital allocation proposal that withstands finance scrutiny. The writer—whether a novelist or a real estate strategist—must understand the audience, the structure, and the stakes. In this context, just as a novelist structures a plot, a strategist must structure a business case that turns raw data into a board-level argument, a process that narrative structuring tools like the how Unsloppy AI Novel Writing App fits the writing workflow can support by helping refine structure, though the strategic thinking remains a distinctly human capability.
The chiller got funded because the case was legible. The analytics platform got tabled because the case was not. The technology gap is closing. The narrative gap is still wide open. Close it, and you change which investments your organization makes—and which futures it bets on.
Margaret Sinclair is a corporate strategy analyst at Brookfield Johnson Controls. She writes about the intersection of real estate, building technology, and corporate operations, helping organizations make better decisions about the physical infrastructure that shapes their future.