Executive Summary: U.S. commercial real estate owners are losing an estimated $15 billion a year to CAM and operating expense reconciliation errors, and roughly 40% of reconciliations contain a material mistake. Most of that leakage isn't fraud — it's a data problem: teams reconcile against summary spreadsheets instead of the actual lease provisions governing caps, exclusions, and gross-ups. As portfolios grow more complex and tenants get more sophisticated about auditing their bills, the owners who win are the ones who catch these errors before the invoice goes out, not after a tenant's auditor finds them. That requires knowing, at the lease level, exactly what every document allows — which is a data problem before it's an accounting problem.
Every CAM season, property accounting teams run the same process: pull actual operating expenses, allocate them across the rent roll by pro rata share, apply whatever caps and gross-ups the leases specify, and send the bill. The math is straightforward. The inputs are not.
Tango Analytics has found material errors in roughly 40% of commercial CAM reconciliations. PredictAP estimates that unrecovered CAM overcharges — money landlords bill incorrectly, tenants catch, and eventually claw back — total around $15 billion annually across U.S. commercial real estate. Roughly one in five tenants encounters double billing in their CAM statements at some point in the lease term.
Those aren't rounding errors. On a mid-size office or retail portfolio, a few percentage points of CAM misallocation is real money moving in the wrong direction — either overbilling tenants (inviting disputes, audits, and litigation) or underbilling them (leaving recoverable expenses on the table, which is the quieter but equally damaging version of the same problem).
The recurring sources of CAM reconciliation errors are well known, and none of them are exotic:
Individually, each of these is a known, fixable error. At portfolio scale, across hundreds of leases with amendments layered on top of amendments, they compound into the $15 billion problem.
Here's the part that gets missed: most reconciliation failures aren't accounting failures. They're data failures upstream of the accounting. The reconciliation software did the math correctly — on the wrong inputs, because the CAM cap, the exclusion list, or the gross-up methodology that got entered into the system months or years ago didn't match what the lease actually says.
This is the same structural problem showing up across every corner of CRE data right now. PwC's Emerging Trends in Real Estate 2026 names data infrastructure quality as a top-three strategic differentiator for CRE firms this year, and notes that the gap between firms with unified, verified data layers and those still reconciling across spreadsheets is widening with every deal cycle. CAM reconciliation is simply the line item where that gap shows up first, because it happens every single year, on every single lease, whether or not anyone is paying close attention.
Lease documents are the ultimate source of truth here — not the abstract summary, not the spreadsheet built off last year's abstract, not the property manager's memory of "how we've always done it" for that tenant. If the CAM cap language, the exclusion list, or the gross-up mechanics captured in your system don't match the actual executed lease and its amendments, the reconciliation is wrong before anyone runs a single calculation.
The obvious fix — have someone re-read every lease before reconciliation season — doesn't survive contact with a real portfolio. A 300-property portfolio with an average of four amendments per lease means thousands of documents, each with its own CAM structure, that would need re-verification every year against changing operating expense actuals. No property accounting team is resourced to do that manually and still hit reconciliation deadlines.
This is precisely the kind of problem AI for CRE is suited to, provided the AI is actually reading the source documents rather than inheriting whatever was typed into a system three lease administrators ago. Prophia was built around that distinction: AI-powered extraction paired with human validation, applied directly to the lease and its amendments, so the CAM cap, exclusion list, and gross-up provision driving this year's reconciliation are pulled from — and checked against — the actual document, not a legacy summary. Across the 650M+ square feet Prophia represents, that verification step has surfaced millions of dollars in lease-level discrepancies that would otherwise have flowed straight into a reconciliation, an audit dispute, or a lost recovery.
Tenants have gotten more sophisticated about this. Professional CAM audit firms routinely recover 3–5% of a tenant's annual occupancy costs, and that industry exists specifically because owners' reconciliation data doesn't match what the lease allows. Every audit a tenant wins is, definitionally, an error the owner's own systems should have caught first.
The strategic implication for asset managers and property accountants is straightforward: the owners who get ahead of CAM accuracy — verifying cap structures, exclusions, and gross-up methodology against the actual lease before the bill goes out — stop funding the tenant-side audit industry with their own reconciliation errors. That's not just a compliance win; it protects NOI and removes a recurring source of tenant disputes that erodes the relationship at renewal time.
CAM reconciliation is a preview of where CRE data accuracy is going more broadly. As AI adoption in property accounting and lease administration accelerates — and industry surveys already show the majority of owners piloting AI tools somewhere in their operation — the organizations that benefit are the ones whose AI is reading verified, structured lease data rather than legacy spreadsheets. An AI model that reconciles CAM against the wrong cap just automates the same error faster.
The next 12–18 months will separate two camps: owners who treat lease data as a one-time abstraction exercise, and owners who treat it as a continuously verified foundation that every downstream process — reconciliation, budgeting, audit response, disposition underwriting — draws from. The second camp is the one that stops losing money to errors nobody was looking for.
Owners who want to see what their portfolio's CAM caps, exclusions, and gross-up provisions actually say — verified against the source documents, not last year's summary — can see how Prophia's customers are approaching this ahead of the next reconciliation cycle, or request a demo directly.