A structural engineer can retire, close the firm, and sell the house — and still get sued eleven years later over a roof that collapsed under snow load. That's not a scare story; it's how professional liability works in this field. Design claims can surface long after a project is "done," which is exactly why a structural engineering firm's books need to track more than the usual accounts receivable and payroll. They need to track risk that doesn't expire when the invoice gets paid.
Structural engineering firms sit in an unusual spot in the professional-services world. The work looks like consulting — hourly rates, project fees, a small team of licensed professionals — but the liability profile looks like construction, and the revenue-recognition mechanics look like a general contractor's. Getting the bookkeeping wrong in any one of those three dimensions creates problems that don't show up until a bank asks for financials, a partner asks for a buyout valuation, or a claim lands on a project nobody's touched in years.
Why "Just Track Hours and Invoices" Doesn't Work Here
Most structural engineering firms take on a mix of project types: a fixed-fee residential retrofit, a percentage-of-construction-cost commercial job, an hourly forensic investigation after a building failure, and a subscription-style retainer for a developer's ongoing portfolio. Each of those needs to be recognized differently, and lumping them into one "professional fees" bucket in your ledger makes it impossible to tell which project types are actually profitable.
The three things that make this industry's books distinctive are:
- Long project timelines with milestone-based deliverables (not physical construction progress)
- Claims-made insurance that creates liability years after the cash has already been collected and spent
- A licensed professional's stamp as a distinct, billable — and sometimes uninsurable — asset
Each deserves its own approach.
Percentage-of-Completion, But on a Drawing Set, Not a Building
Percentage-of-completion (PoC) accounting is standard for long-duration contracts under GAAP, and it's the right method for most structural engineering fee contracts that run more than a few weeks. But contractors measure completion by physical progress — cubic yards poured, square footage framed. An engineering firm has no physical structure to measure against. What you're actually completing is a stamped drawing set, and that has to be estimated by design phase, not by percentage of a building that isn't yours to build.
The standard phase breakdown most structural firms use for PoC estimates:
- Schematic design / conceptual framing scheme — roughly 15–20% of the fee
- Design development (member sizing, lateral system selection) — another 25–30%
- Construction documents (the drawings that actually get stamped and issued for permit) — the biggest chunk, often 35–40%
- Construction administration (RFIs, submittal review, site visits during building) — the remaining 10–15%, often billed hourly against a not-to-exceed cap
A few bookkeeping consequences follow directly from this:
- Revenue recognized should track phase completion, not calendar time or cash received. A firm that's 60% through construction documents on a six-month contract should recognize roughly 45–50% of the total fee (schematic + design development + a chunk of CDs), not 50% just because three months have passed.
- The stamp-and-seal moment is a billing trigger, not a revenue-recognition trigger. Many firms bill a discrete "PE stamp and seal" line item (commonly $100–$300 per sheet, sometimes bundled into the CD-phase fee) at the point drawings are issued for permit. That's a cash-collection event tied to a specific deliverable, and it should hit accounts receivable as a milestone invoice — but revenue was likely already partially recognized in prior periods under PoC if the firm is doing this correctly. Reconcile the two rather than letting billing drive your P&L.
- Construction administration hours need their own cost code. CA work often spans many months at low hourly volume, and it's easy for a project manager to eat the cost by not tracking it separately from the CD phase that funded most of the fee. If CA isn't broken out in your chart of accounts or job-costing categories, you'll systematically understate how unprofitable long CA tails actually are.
- A firm crossing roughly $32 million in average annual gross receipts (2026, three-year lookback) is required under IRC §460 to use percentage-of-completion for long-term contracts for tax purposes — most independent structural firms are well under this threshold, but it's worth knowing where the line sits if you're scaling or merging with a larger multidisciplinary practice.
Work-in-Process Isn't Just a Report — It's Money You Haven't Billed Yet
Ask most engineering firm owners what their accounts receivable balance is and they'll tell you instantly. Ask what their unbilled WIP is — completed design work sitting between "the phase is done" and "the invoice went out" — and you'll often get a shrug. That gap is real, uncollected revenue, and it's one of the most common reasons a structurally profitable firm shows weak cash flow.
Before each billing cycle, someone should be reviewing every active job and asking: what phase is this actually at, what's been billed so far, and does the gap match? A drawing set that's 80% through construction documents but has only been invoiced through schematic design isn't a rounding error — it's a project manager sitting on a month of billable work because reviewing markups felt more urgent than sending an invoice. Left unmanaged, WIP aging becomes the quiet twin of AR aging: nobody watches it, and it's usually where the cash is actually stuck.
E&O Tail Coverage: The Liability That Outlives the Project
Here's the part that has no equivalent in most other small-business bookkeeping guides. Structural engineering E&O (errors and omissions) insurance is almost always written on a claims-made basis — meaning the policy has to be active when the claim is filed, not when the design work happened. A building designed and stamped this year can generate a claim five, eight, even eleven years later if a structural failure gets traced back to a design decision, and if there's no active policy (or tail coverage) in place at the moment that claim is filed, the firm — or the individual PE who stamped the drawings — can be personally exposed with zero coverage.
Statutes of repose set the outer boundary on how long that exposure lasts, and they vary sharply by state — commonly seven to ten years after substantial completion in most states, but New York and Vermont have no true statute of repose for design-defect claims at all, leaving exposure open-ended. If your firm works across state lines, your longest-exposure state effectively sets your tail-coverage floor.
What this means for the books:
- Tail coverage (an "extended reporting period") typically costs 100–300% of your annual premium, depending on how many years of extension you're buying. This is not a routine renewal line item — it's a significant, lumpy cost that should be planned for, not discovered.
- Budget for it at three trigger points: when a firm dissolves or merges, when a principal retires, and when switching insurers (a new claims-made policy typically won't cover work performed under the old one unless you negotiate "prior acts" coverage or buy tail on the old policy).
- Many client contracts now mandate the firm carry tail coverage for a specified period — commonly five years — after project completion. That's a contractual liability that should be tracked project-by-project, not assumed to be covered by whatever policy happens to be active when the contract is signed.
- Treat the eventual tail-coverage cost as a long-tail liability on the books, similar to how a contractor reserves for warranty work — not a surprise expense the year a partner retires.
The PE Stamp Is an Asset on the P&L, Not Overhead
It's easy to bury a Professional Engineer's licensing costs, continuing education, and — most importantly — their time spent reviewing and sealing drawings inside general "professional development" or "overhead" line items. That undercounts what a stamp is actually worth to the firm.
A few practical adjustments:
- Cost a PE's stamping and QA-review time against the specific project, not as unallocated overhead. If a senior PE spends four hours reviewing a junior engineer's calculations before sealing a set, that's billable (or at minimum job-costable) time tied to that project's margin — not a fixed cost spread evenly across every job the firm touches that month.
- In states with full or partial practice restriction — where only an SE-credentialed engineer can seal "designated structures" like high-rises, hospitals, or schools — track which staff hold which credential and which project types they're eligible to stamp. This isn't just a staffing question; it determines which jobs the firm can even bid on, and losing your one SE-credentialed engineer to a competitor is a capacity loss worth modeling, not just a payroll line that disappears.
- A fixed-fee contract that doesn't separately price the stamp-and-seal line item is quietly absorbing licensing risk into the base fee. If your invoices don't show that line, check whether it's priced into the CD phase at all — a startling number of small firms simply forget to charge for it.
Keep Your Financial Records as Clear as Your Calculations
Structural engineers already think in terms of load paths, safety factors, and documentation that has to hold up under scrutiny years after the fact — the same discipline applies to your books. Beancount.io offers plain-text accounting that's transparent and version-controlled, so every phase-completion adjustment, WIP reclassification, and tail-coverage reserve is documented and auditable, not buried in a spreadsheet formula nobody remembers writing. Get started for free and keep your financial records as rigorous as your calculations.