Marketing Agencies: Cube vs Mosaic for Net Fee Planning

Start the worksheet with two revenue rows, not one: gross billings (everything that flows through your accounts, including client media spend and pass-through production costs) and net fee revenue (what's actually yours after paying media platforms and vendors). Confusing the two is the single most common way an agency's growth story gets overstated to a bank, an investor, or even to the agency's own leadership team.

In Cube, this is a straightforward two-row formula where net fee revenue equals gross billings minus a pass-through cost line you build from vendor and platform invoices. In Mosaic, you'd typically exclude pass-through amounts from the revenue feed entirely at the data-connection stage, so net fee revenue becomes the only number the platform ever sees, which is cleaner but means gross billings has to be tracked somewhere else if you still want to report it to clients.

Retainer clients typically give 30 days notice before leaving, which means your revenue forecast has real, near-term risk baked into it that a typical SaaS churn model, built around annual contracts, doesn't reflect well. Build a row that flags retainers approaching a renewal or review date, separate from the steady-state retainer total, so a client relationship going quiet shows up as a forecast risk before the notice actually arrives.

Mosaic's churn cohort logic can be adapted for this, but it assumes churn happens on a subscription renewal cycle, monthly or annual, rather than on a rolling 30-day notice period tied to relationship health rather than a contract date. Cube lets you build the flagging logic around whatever signal your account team actually uses, like a client going quiet on a monthly report.

Unlike payroll, freelance production cost often gets committed a quarter ahead for a specific campaign, which means it doesn't flex down the moment a retainer client leaves the way agency owners sometimes assume. A worksheet that shows freelance cost as a variable expense tied to current revenue will understate the cash exposure if a client churns mid-quarter after production work is already committed.

Build freelance and contractor commitments as their own row with a lag, showing cost obligations already locked in for the next 60 to 90 days regardless of what happens to revenue in that window. This is the row most agency budgets skip, and it's usually the one that explains a cash surprise after losing a client.

Account and creative staff compensation tends to track broader labor market trends closely, since agency talent isn't a specialized enough pool to command a large premium the way, say, compliance specialists do. Wage growth across the civilian workforce has been running at 3.4% year over year¹, a reasonable floor for your annual merit-increase assumption before layering on any retention raise for a senior account lead you don't want to lose to a competing agency.

• Net fee revenue, not gross billings, should be the number every other ratio in your model is built against, including any margin or headcount-to-revenue calculation.
• Flag retainers approaching a renewal or review window as a distinct forecast risk, not folded into steady-state revenue.
• Show freelance and production commitments with their actual lag, not as a variable cost that flexes instantly with revenue.
• Build payroll growth off a realistic wage assumption, then add retention premiums for specific people you'd genuinely struggle to replace.

Either tool can hold this worksheet once you've defined the rows. The harder work, and the part that actually matters, is agreeing on the gross-to-net split and the churn-risk definition before you build anything.