Tenant Guide: Auditing a US Commercial CAM Reconciliation

A CAM reconciliation is the annual statement a US commercial landlord sends each tenant comparing the year's actual common-area operating expenses against the monthly estimates the tenant paid, with a true-up balance owed in either direction. On a fully-reconciled NNN lease, the same cycle covers real estate taxes and insurance on parallel statements; "CAM" is used here as shorthand for the whole reconciled pass-through, not just common-area maintenance in the narrow sense. Landlords typically deliver the statement 90 to 120 days after fiscal year end. Tenants then have between 90 days and 12 months from delivery — the audit window the lease defines — to review the statement, request supporting invoices, and dispute discrepancies. Once the audit-request letter goes out, most leases give the landlord around 30 days to produce the supporting GL extract and invoice package.

Findings on a US commercial CAM reconciliation tenant audit cluster in three patterns: capital expenditures misclassified as operating expenses, management fees applied to a grossed-up base that exceeds the lease's contract cap, and gross-up calculations applied to expenses that shouldn't vary by occupancy. Other findings appear, but those three categories drive most of the disputed dollars in most audits.

The reconciliation statement is structured around the lease's CAM clause, which is the controlling document throughout. The clause defines what's in the operating-expense pool and what isn't, sets caps on the management fee and any administrative fee, governs whether expenses can be grossed up to a stated occupancy and which expenses qualify, and grants the audit right itself. Every line on the statement either falls inside that definition or doesn't, and that determination — not industry custom or the landlord's own categorisation — is what the audit tests. The contents of the pool itself originate as the landlord-side AP workflow that builds the same expense pool a tenant later audits, which is why the audit work is essentially reverse-engineering the landlord's AP coding decisions back through the lease.

This guide is the audit-execution workflow, from envelope-arrival to findings letter. It assumes the reader already knows what their pro-rata share is and what their lease's audit-rights clause says. It is not a lease-clause drafting guide and it is not legal advice; it's how a tenant accountant, in-house real-estate operations lead, outside CAM auditor, or CPA actually grinds through a 50-to-500-page backup-invoice package and produces a defensible findings letter at the end.

The desk audit before you request backup

Before the audit-request letter goes out, the reconciliation statement on its own already supports a desk review that decides which categories are worth chasing. The point isn't to resolve findings without backup — most categories can't be resolved that way — it's to produce a prioritised list of categories and line items so the request letter is targeted and the production-deadline clock isn't burned on a blanket ask.

Start with the bottom-line arithmetic. Actual operating expenses minus estimates collected from the tenant during the year equals the true-up balance, signed correctly. This is the single most basic check on the statement, and it's occasionally wrong — usually when the prior-year estimate was reset mid-year and the statement compares against the wrong estimate base. If the math doesn't tie, the rest of the statement is suspect.

Next, verify the pro-rata share number on the statement against the lease. The numerator is the tenant's leasable area; the denominator is the building's leasable area for the period. If the building's leasable area changed mid-year — a new wing came online, a portion went offline for renovation, an existing tenant expanded — the share denominator should reflect a weighted average over the reconciliation period, not a single year-end snapshot. Flag mid-year changes as items requiring closer attention later; this is where pro-rata errors compound across categories.

Spot-check the share applied to each major category. Some leases use different denominators for different categories: a retail-only category (marketing fund contributions, signage maintenance) might pool only across retail tenants while a building-wide category (security, common-area utilities) pools across the whole building. A statement that applies a single share across every category is a common error pattern, and the categories that benefit from the wrong denominator depend on whether the tenant is mostly in the retail pool or the building pool.

Read the management-fee and administrative-fee lines against the lease's stated caps. If the lease caps the management fee at four percent of operating expenses, divide the reported management-fee dollar by the operating-expense base on the statement and check whether the implied percentage matches. If the cap is on actual expenses but the fee was applied to a grossed-up base, the implied percentage on actuals will be over the cap. Same arithmetic on any administrative-fee cap, whether the lease states it as a flat dollar amount or a percentage.

Look for any indication the expense base has been grossed up. Gross-up is usually disclosed in a footnote on the statement or in a separate worksheet line; sometimes it appears as a "stabilised expenses" or "gross-up adjustment" line item. If the statement shows a gross-up but the lease doesn't permit one, the entire statement is structurally wrong and not just a category-level finding. If gross-up is permitted, note the occupancy percentage used and whether it matches the lease's stated percentage.

Compare the year's totals against the prior year and against the current year's estimates. Any category that swung more than ten percent year-over-year — or moved by a meaningful dollar threshold relative to the property's overall operating budget — becomes a candidate for backup-invoice review. Significant overruns versus the year's own estimates are a separate flag: a category that came in 30 percent above budget without an obvious driver (a known capital project, a known service-contract renewal) usually means the landlord's budgeting process missed something the audit will surface.

Scan for line items that are exclusion-eligible on their face. Capital improvements appearing in routine maintenance lines, marketing or promotional-fund contributions inside CAM, landlord-principal compensation tucked into management or administrative lines — the lease's exclusion list is the controlling reference, and any line that names something the lease excludes is a flag without needing backup at all.

The output of the desk audit is the prioritised list: categories where the statement on its own raised a question, plus categories large enough that line-item review is worth the effort regardless.

Requesting the supporting invoice package

The audit-request letter is what triggers the lease's production-deadline clock — commonly around 30 days, sometimes longer for fully NNN portfolios where the package is large. Send it once the desk audit has produced its prioritised list, not before. A blanket request sent on receipt of the statement burns the clock on a package that mostly turns out not to be in dispute.

The request should ask for the GL extract for the entire CAM expense pool, not just the categories the desk audit flagged. The extract is the bridge between the summary statement and the supporting invoices: it lists every expense the landlord booked into operating expenses for the year, by GL code, with vendor and amount. Without the full extract, the audit can only verify expenses the landlord chose to put on the statement; with it, the audit can also see what the landlord excluded — which sometimes turns out to be expenses that should have been included as a credit (vendor rebates, insurance refunds, parking-lot revenue) but weren't.

For invoice-level backup, the practitioner threshold is commonly line items over $5,000, with all line items inside flagged categories regardless of dollar amount. Some leases set their own threshold (often $10,000 or a percentage of operating expenses); where the lease specifies, the lease controls. Where it doesn't, $5,000 is the working norm. Smaller line items can usually be sampled rather than fully reviewed.

Two specific worksheets are worth asking for explicitly because they almost never appear without being requested. The management-fee calculation worksheet should show the base figure the fee was applied to, the percentage applied, and any adjustments — this is what makes a fee-cap audit possible without reverse-engineering the math. The gross-up worksheet should show which expense lines were grossed up, the actual occupancy used, the target occupancy the gross-up scaled to, and the resulting adjustment per line. Without the worksheet, gross-up errors hide inside aggregate category totals.

Service contracts behind every recurring vendor expense are the third request category. Janitorial, snow removal, security, landscaping, elevator service, HVAC preventive maintenance — each typically runs on a multi-year contract, and the contract establishes whether the invoiced amount matches the contracted scope, whether mid-year scope changes were properly authorised, and whether the contracted services are CAM-eligible under the lease at all. A janitorial contract that includes cleaning of the landlord's leasing office, for example, is a contract-level finding that no individual invoice review would surface.

What arrives back is rarely tidy. The package is typically a 50-to-500-page PDF mixing the GL extract (often as the first 10 to 30 pages, exported from Yardi or MRI), per-vendor invoice scans grouped by category or by month, the service contracts as separate sub-PDFs, the management-fee and gross-up worksheets as one or two trailing pages, and sometimes scanned receipts where the landlord's AP function captured something below the threshold for a vendor invoice. Native-PDF invoices and scanned ones sit side by side in the same package; the same vendor might appear three times in three different formats depending on how the AP team processed each month's bill.

Some landlords deliver the package as a ZIP of PDFs organised by category, which is the easiest version to process. Some send a single concatenated PDF, which is harder. A few — usually older property-management operations — still require in-person review at the management office, with the package available as a printed binder you can photograph or scan but not take. The package's shape reflects how the landlord's AP function is set up, which is why it varies so widely from landlord to landlord even within the same market and the same property type.

Turning the backup invoice package into an audit workbook

The package is now on your desk. The audit doesn't start until it stops being a PDF and becomes a workbook.

Concretely, the conversion is from a 50-to-500-page mixed PDF — GL extract from Yardi or MRI, per-vendor invoice scans, service contracts, management-fee worksheet, gross-up worksheet, sometimes a tail of scanned receipts — into a structured workbook where each row is one supporting invoice or one GL line. The columns that make the rest of the audit tractable are: vendor name, invoice date, invoice amount, GL category as the landlord coded it, lease-clause check (in, out, or partial under the lease's operating-expense definition), allocation method (direct to common areas, pro-rata across all tenants, or specific allocation to a sub-pool), this-year amount, prior-year amount for the same vendor or category, percent variance, dispute flag, and a supporting-evidence reference back to the PDF page or filename.

This schema isn't optional. Without it, the audit collapses into ad-hoc note-taking — sticky tabs on the PDF, a side document of "things that look weird," running totals that don't reconcile to the statement — and the findings letter at the end can't cite line-item evidence that holds up. With it, the next two passes (per-category logic and year-over-year comparison) become filtering and sort operations on a populated workbook, not full re-reads of the package.

The honest part is that converting a 500-page mixed PDF into that workbook by hand is the time sink in every CAM audit. A practitioner reading line by line, transcribing vendor and amount and date and GL code into Excel, is doing five to fifteen hours of mechanical extraction work before any audit judgment begins. Every guide and every audit-services landing page on the SERP treats this step as a footnote — "request invoices over $5,000," "trace significant expenses to original invoices" — but the practitioner reality is that the trace is most of the work and the audit judgment is the smaller share.

Which is also why this is the natural place for automated extraction of the landlord's backup invoice package for a CAM tenant audit to land. Invoice Data Extraction takes the landlord's PDF — native-PDF and scanned invoices mixed, with the GL extract, service contracts, and worksheets all bundled in — and converts it into a structured spreadsheet using a single natural-language prompt that describes the audit workbook schema: vendor, invoice date, amount, GL category, allocation, supporting-evidence reference. The output lands as Excel, CSV, or JSON, and every row references the source page in the original PDF, which is the same supporting-evidence pointer the audit needs in the workbook anyway. A 500-page CAM backup package, or a ZIP of separate per-category PDFs, both fit inside standard processing.

Category-by-category audit logic

With the workbook populated, the per-category review is what most of the audit's substantive work consists of. The structure for each category is the same: name the lease-clause check first (does the lease's operating-expense definition include or exclude this expense, and on what basis), then the typical discrepancy pattern, then the supporting evidence that confirms or refutes the line.

Janitorial. Almost always included as a CAM-eligible expense for common-area cleaning. The discrepancies surface where the cleaning crosses out of common areas: cleaning of tenant-exclusive premises billed back to CAM rather than to the specific tenant, cleaning supplies for the landlord's own management or leasing offices folded into the janitorial line, contracted scope expanding mid-year (additional weekly visits, post-COVID disinfection add-ons that became permanent) without the rate change being separately disclosed. Supporting evidence is the cleaning vendor's service contract, the monthly invoices, and any scope-of-work document if the line stepped up year-over-year.

Trash and waste removal. Common-area waste — building trash compactors, common-area recycling — is included. Tenant-specific waste, particularly anything specialised, is usually billed direct: a restaurant tenant's grease-trap servicing, a medical tenant's biohazard pickup. Discrepancy patterns are tenant-specific waste pulled into the common pool, and fuel surcharges or environmental fees that inflate the per-pickup cost without changing the contracted scope. Hauler invoices and the contract showing pickup frequency are the evidence.

Security. Common-area security is included; security at the landlord's leasing office, management office, or any landlord-occupied space is excluded. Capital security upgrades — new camera systems, replacement of access-control hardware, fence installations — are usually capital under the lease, not operating, even when the line item reads as "security services." A camera-system replacement billed inside a monthly security line is a common pattern.

Snow removal and landscaping. Both included as common-area maintenance under most leases. The discrepancies are scope-related: extreme-weather event responses billed at premium rates without the contract specifying the premium, landscape redesign or replanting (capital under most leases) billed as routine maintenance, and seasonal expenses double-counted across overlapping invoices. Snow removal in particular invites disputes when the contract is per-event and a single storm produces three sequential service invoices.

Common-area utilities. Utilities serving common areas — parking-lot lighting, common-corridor HVAC, exterior building lighting, common-area water — are typically included. Tenant-metered utilities are not; utilities for landlord-occupied space are not; and utilities for vacant tenant space are usually treated under the gross-up provision rather than billed at full cost. The discrepancy patterns are landlord-occupied or vacant-space utilities pulled into the common pool, and meter-allocation errors when the property has multiple sub-meters that don't cleanly map to common versus exclusive areas.

Elevator and HVAC maintenance. This is one of the cleanest capital-versus-operating boundaries in the package, and it's also where the largest single-line errors usually hide. Routine maintenance — preventive maintenance contracts, routine service calls, parts replacements under the contract — is operating. Major component replacement is capital under most leases: an HVAC compressor or full unit replacement, an elevator modernisation, a chiller replacement. The pattern is for these capital items to be billed inside the recurring maintenance line under an innocuous description like "HVAC service" or "elevator repair," sometimes split across two or three monthly invoices to obscure the unit cost. The supporting invoice tells you which it actually was; the contractor's scope language settles it where the invoice description is ambiguous.

Management and administrative fees. Both are capped, both have their own arithmetic, and they're usually negotiated together but separately. The management fee is commonly four to five percent of operating expenses on NNN properties. The administrative fee, where it appears, is typically a flat dollar amount or a smaller percentage (one to two percent). For the management fee, the discrepancies are: applied to a grossed-up base when the lease caps it on actual expenses, computed on a base that includes excluded items (some leases exclude taxes and insurance from the management-fee base), or computed at a percentage above the contracted cap. For the administrative fee: stacked on top of the management fee in a way the lease does not authorise — some leases explicitly prohibit doubling — or computed against the grossed-up base when the lease ties it to actuals. The management-fee calculation worksheet is the evidence for both, with the base figure and the percentage applied visible side by side.

Insurance and real estate tax allocations behave like specialised CAM categories with their own audit considerations. Insurance: the lease defines which policies are CAM-allocable. Building property insurance and general liability for common areas are typically included; umbrella policies, landlord-specific coverage (D&O, errors-and-omissions for the management entity), and any insurance covering specifically tenant-leased space are typically not. The discrepancies are pro-rata share applied to policies that should be allocated only to specific sub-pools, and umbrella premiums included at full value rather than at the building-attributable share. Real estate tax: on a fully NNN lease, tax is usually a separate reconciliation, but the audit logic is identical. Watch for tax abatements or refunds the landlord received but didn't flow through to the tenant, and special assessments — for street improvements, district levies, or similar — that many leases exclude as capital-natured impositions.

The recurring discrepancies that drive most findings

The per-category review surfaces category-specific issues. Cross-cutting patterns surface across categories, and these are where most of the disputed dollars actually sit.

Capital expenditures misclassified as operating expenses. The single largest dispute by dollar value in most audits, because capital items are usually the largest individual invoices in the package. Roof replacement, HVAC compressor or unit replacement, parking-lot resurfacing or repaving, elevator modernisation, parking-structure structural repairs — all capital under most commercial leases, and all routinely categorised as operating in landlord billing. Sometimes the categorisation is willful; sometimes it reflects how the landlord's AP team coded the invoice without anyone applying the lease's operating-expense definition. Either way, the lease is the controlling reference. Supporting evidence is the underlying vendor invoice and, where the work was substantial, the project scope document or service contract that establishes whether the work was a replacement (capital) or a repair (operating).

Management fees applied to a grossed-up base. The discrepancy is two layers deep. First: was the gross-up itself permitted under the lease, and was it applied correctly. Second: was the management-fee percentage then computed against the grossed-up base in a way that exceeds the lease's stated cap on actual expenses. The arithmetic example is direct. A four percent management fee on actual operating expenses of $1.0 million is $40,000. The same four percent on a grossed-up base of $1.1 million is $44,000. If the lease caps the fee on actual expenses, the additional $4,000 is overbilled — and that's only the per-year amount. The same pattern repeating across a five-year audit window stacks up.

Gross-up calculations applied to fixed costs. Gross-up exists to prevent the landlord from absorbing the variable-occupancy share of expenses when the building isn't fully leased — it is not a general inflation mechanism, and it isn't supposed to apply to expenses that don't change with occupancy. The authority on this is unambiguous: how commercial-lease gross-up provisions distinguish variable from fixed operating expenses is direct that commercial-lease gross-up provisions let a landlord overstate variable operating expenses as if the building were 95% or 100% occupied (the exact percentage is negotiated by the parties), and only expenses that vary by occupancy — such as electricity, utilities, trash removal, management fees, and janitorial services — should be grossed up; expenses that do not vary by occupancy, including taxes, insurance, and building security, should not be. A gross-up applied across the entire expense pool, rather than only to the variable subset, is a structural error that recurs in every category that shouldn't be in scope.

Administrative fee exceeding the lease's stated cap. The administrative fee is usually negotiated separately from the management fee, with its own cap structure: a flat dollar amount, a smaller percentage, or a combination. The discrepancy is the same shape as the management-fee version — applied to a base the lease doesn't permit, or computed at a percentage above the cap. Audit the administrative-fee dollar against the lease's cap formula directly; don't assume the cap structure mirrors the management fee.

Marketing or promotional-fund contributions categorised as CAM. Common in retail leases where the lease structures marketing as a separate fund with its own contribution formula and cap. Pulling marketing contributions into the CAM pool inflates two things at once: the line itself, and the management-fee base if the management fee is computed on operating expenses inclusive of the misallocated marketing dollars.

Landlord-principal compensation included in operating expenses. Some landlords include the principal's salary, owner's draw, or related-party payments to a management entity owned by the same principal within management or administrative line items. Most leases exclude this either explicitly or through the management-fee cap structure, which is intended to be the entire compensation for managing the property. Related-party payments are worth verifying explicitly in the GL extract — same surname, same address, suspiciously round dollar amounts.

Pro-rata-share calculation errors when leasable area changed mid-year. New tenants moving in, existing tenants expanding or contracting, a portion of the building taken offline for renovation, the conversion of common space to leasable space (or vice versa) — all change the denominator. A static share applied across the year, when the underlying area changed, produces small per-category errors that aggregate to material amounts. Mid-year area changes are also where landlord and tenant interpretations of the share calculation tend to diverge most.

The capital-versus-operating boundary deserves particular attention because reclassifying a single capital item back out of operating expenses often produces a larger adjustment than the rest of the audit's findings combined. A $200,000 roof replacement billed as operating expense, on a tenant with a 10 percent pro-rata share, is a $20,000 finding before the management fee on the inflated base is even unwound.

Year-over-year comparison as the outlier-detection step

The per-category and discrepancy-pattern reviews catch what the lease's text and audit experience tell you to look for. Year-over-year comparison catches the rest — re-categorisations that are invisible inside a single year, contract step-ups the landlord didn't separately disclose, new line items that don't fit anywhere obvious, and silent rolling of expenses across category boundaries.

The structure is straightforward. Categories as rows, years as columns: this year, prior year, ideally one or two further back. Two extra columns: percent change year-over-year, dollar change year-over-year. Three years of data is usually enough; longer horizons risk comparing across operating regimes (a vendor change, a service-contract restructure, a property renovation) where the comparison stops being meaningful.

The filter is threshold-based. Any category with a year-over-year change above the threshold gets flagged for closer review. The most common threshold is ten percent change, sometimes paired with a dollar floor — a category that grew 12 percent but only $200 isn't worth chasing on a $20 million expense pool, and a category that grew 6 percent but $80,000 is. The right threshold scales to the property and the audit's materiality.

For each flagged category, the investigation runs through the workbook's supporting-evidence references. The questions are concrete: what's a new vendor on this year's package that wasn't there last year? What contract scope changed? Did the same vendor change rates? Did service frequency change — more visits, more pickups, more service calls? Did the same dollar of expense move from one GL category to another, leaving the source category looking smaller and the destination category looking larger? The workbook makes these tractable; without it, the same questions would require flipping between this year's PDF package and last year's.

A handful of patterns are worth investigating specifically when a delta crosses the threshold:

• A new line item this year that didn't appear last year is usually either a category re-classification (the same expense moved across a GL boundary) or a new pass-through the lease may not authorise.
• A line item that disappeared this year is sometimes legitimate (contract ended, vendor changed, service discontinued) and sometimes a roll-up — the same expense booked into a different category, often a less-scrutinised one.
• A category that grew faster than CPI without an obvious driver is usually a contract step-up the landlord did not call out separately. Service contracts commonly have annual escalators, but a step-up beyond the contractual escalator suggests something else.
• A category that grew exactly in line with the management-fee base, rather than with its own underlying drivers, is sometimes an artifact of how the gross-up was applied across the pool — the variable-and-fixed distinction collapsed at the spreadsheet stage.

Comparison against the current year's estimates is a separate axis worth running. Significant overruns versus the year's own budget signal a budgeting failure on the landlord's side; if the overrun pattern is recurring (the prior year overran in the same categories), it justifies a request that the next-year estimate be reset rather than carrying forward an estimate the landlord and tenant both know is wrong.

The technique has limits. Year-over-year comparison surfaces outliers; it doesn't on its own produce findings. Each flagged category still gets resolved through the per-category lease check and the discrepancy-pattern logic from the prior sections. What the comparison adds is the confidence that nothing material was missed because it sat inside an expected category at an unexpected size.

Producing the audit findings letter and feeding the downstream

The audit's output is two artifacts: the workbook, and the findings letter the workbook supports.

The completed audit workbook is the supporting record. Categories as rows, years as columns, with the discrepancy-flag column populated and the supporting-evidence-reference column pointing back to the specific PDF page or filename for every disputed line. The workbook is what gets exchanged with the landlord during dispute resolution — the landlord's auditors or counsel will work from the workbook the tenant produces, not from the findings letter alone, and a workbook that doesn't tie back cleanly to the original package weakens the tenant's position in the back-and-forth.

The line-item discrepancy format inside the findings letter follows a consistent shape. For each disputed item: the lease clause cited (specific section reference, not just "the audit-rights clause"), the landlord's billing as it appeared on the statement, the tenant's position with the corrected amount, the dollar impact at both the gross level and the tenant's pro-rata share level, and the supporting-evidence reference. Without the lease clause and the supporting invoice reference, a finding becomes a negotiating position rather than a substantiated claim, and the difference shows in the response the landlord sends back.

The findings letter itself is short. A cover paragraph names the audit period, cites the lease's audit-rights clause, and frames the basis on which findings are being raised (review of the reconciliation statement, the GL extract, the supporting invoice package, and the management-fee and gross-up worksheets). The discrepancy schedule — drawn from the workbook — is the body of the letter, structured as a table or a numbered list of findings with the discrepancy-format detail per finding. A calculated total dollar impact, both gross and at the tenant's pro-rata share, closes the schedule. A request for the landlord's response within the lease's response window closes the letter. Avoid building a long narrative argument; the substantiation lives inside the discrepancy schedule.

The cleaned numbers feed several downstream consumers. ASC 842 treats CAM pass-through as a variable lease payment, so any settled true-up adjustment changes the lease expense recognised in the period it's settled. The audit's net result, after the landlord's response, is what the tenant's lease accounting actually books — which is why the audit work has to be defensible enough to ground a posted journal entry, not just persuasive enough to start a negotiation.

Audit findings should also feed the next year's estimate. A landlord whose estimates have systematically overrun, or whose audit revealed pattern errors (gross-up misapplication, fee-cap drift, capital items routinely classified as operating), should be asked to reset the estimate base rather than carrying the same pattern forward. The audit's leverage on next-year estimates often outlasts the dollar value of the current-year findings.

Findings carry into lease renewal and expansion conversations on the same property too. What the tenant has learned about how this landlord categorises expenses — which clauses got tested, which ambiguities the landlord exploited, which exclusions held up — is renewal-negotiation material. The next lease can be tightened where the current one was loose. This isn't a substitute for a lease-negotiation guide, but it is the most concrete input a renewal team can have.

For tenants whose audit period spans a mid-year change in property ownership, the closing-date split adds an extra reconciliation: what the prior owner billed for their portion of the year, and what the new owner is now billing for theirs. The closing settlement statement is usually the source document that establishes the split, and extracting an ALTA settlement statement into Excel when the lease year begins mid-period from an acquisition is the upstream step that gets the closing numbers into a usable shape before they reach the CAM audit. Where the tenant's organisation also handles owner-side reporting on properties it controls — common in family offices and real-estate operators that sit on both sides of various leases — the parallel discipline of consolidating AppFolio, Buildium, or Yardi owner statements into a portfolio spreadsheet is a useful adjacent reference, because the data-shape problem rhymes even though the audit perspective doesn't.