Engineering Firms: Cube vs Mosaic for Percent Complete

Under percentage-of-completion accounting, revenue recognized equals costs incurred to date divided by total estimated project cost, multiplied by the contract value, which means an updated cost-to-complete estimate directly changes recognized revenue even with no new invoicing. A project manager's optimism or caution about remaining scope has real, immediate financial statement impact.

If your firm needs the flexibility to build this calculation with project-specific adjustments, change orders not yet approved, disputed scope, subconsultant delays, Cube's formula-level access is the better fit. If your projects are relatively standardized and a percent-complete template covers most of your book without much customization, Mosaic's configured approach can save setup time, though this specific accounting method isn't its primary design focus the way SaaS ARR is.

A percent-complete model is only as good as its update cadence. If project managers update cost-to-complete estimates monthly at best, and inconsistently at that, the resulting forecast will lag reality by weeks even with a well-built model. This is a process problem before it's a software problem.

Whichever tool you choose, pair it with a firm policy requiring every active project's completion estimate to be reviewed at least monthly, ideally tied to the billing cycle so the update habit is reinforced by something the project manager already has to do anyway.

Municipal and public infrastructure work often spans multiple fiscal years with funding released in phases, which means backlog isn't a simple sum of remaining contract value, some of it depends on a future budget approval that hasn't happened yet. Treating unfunded future phases the same as signed, funded backlog overstates how much revenue is actually secured.

Split backlog into funded and unfunded categories in your model, and only forecast the funded portion with confidence. Cube's spreadsheet flexibility makes this split easy to build and adjust project by project. Mosaic's backlog and pipeline views tend to assume a more binary won-or-not-won status that doesn't map as cleanly onto a phased public funding structure.

A subconsultant's invoice for work performed in March might not arrive until May, which means a percent-complete calculation based on costs incurred to date will understate actual project costs, and therefore overstate margin, until that invoice lands. This is a common source of a margin surprise at year-end audit that a monthly forecast never caught.

Build an accrual estimate for subconsultant costs incurred but not yet invoiced into your percent-complete calculation, based on the subconsultant's known scope and typical invoicing lag. Either tool can carry this accrual, but someone has to estimate it each month, which is process discipline the software itself won't provide.

Choose Cube if your projects are complex enough that percent-complete calculations need project-specific judgment calls (disputed change orders, phased municipal funding, irregular subconsultant timing) that a standardized template would force into an approximation. Choose Mosaic if your project types are fairly consistent and you'd rather configure a system once than maintain formulas across dozens of active projects. Either way, the firm's estimate-update discipline will matter more than the software choice.

A completion estimate that changes recognized revenue is effectively a financial statement decision, not just a project management update, which means it should go through some sign-off beyond the project manager who proposed it. Firms that skip this step sometimes find a single overly optimistic estimate on a large project inflated a quarter's revenue in a way that then has to be walked back the following period, an uncomfortable conversation with ownership or a bonding company.

Route any completion estimate change above a set threshold, a meaningful swing in either direction, through a principal or controller review before it flows into the forecast. This is a firm policy decision, not a software feature, but it matters more than which platform holds the underlying calculation, since neither Cube nor Mosaic will catch an estimate that's simply wrong.

Confirm these controls are in place before trusting the forecast:

• Require every active project manager to update cost-to-complete estimates on a set cadence, since a lagging estimate leaves the forecast weeks behind reality.
• Split backlog between signed, funded work and unfunded future phases of multi-year municipal contracts, so secured revenue is not overstated.
• Accrue an estimate for subconsultant cost that has been incurred but not yet invoiced, so margin is not overstated until the invoice lands.
• Route any revised estimate that changes recognized revenue through a sign-off beyond the project manager who proposed it.