Cube vs. Mosaic for an Asset-Based Lender's Portfolio Yield
For a specialty asset-based lender, Cube fits a controller who models the portfolio in a spreadsheet and Mosaic fits a book large enough to need a by-vintage dashboard. Either tool has to track portfolio yield net of credit losses, because gross interest income keeps accruing on delinquent loans and hides a credit problem until it becomes a charge-off.
Here's how that plays out, and where Cube and Mosaic each help once portfolio yield and credit loss reserving are the core forecasting problem.
Vendors Covered in this Article
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Walking through why gross yield can hide a real credit problem
Interest income accrues on a loan's outstanding balance regardless of whether that borrower is actually current on payments, at least until a loan is formally classified as non-performing. A forecast that reports gross interest income without tracking accruing-but-delinquent balances separately will show healthy revenue right up until a wave of charge-offs forces a sudden reserve adjustment.
Build the model to track interest income, delinquency status, and reserve adequacy together by loan vintage, so a developing credit problem in a specific origination period shows up early rather than getting averaged into the overall portfolio's still-healthy numbers.
Cost of capital moves with the rate environment, and so should your margin forecast
A specialty lender's own cost of funds, whether from a credit facility, warehouse line, or deposits, typically floats with the broader rate environment, which means portfolio net yield depends on both what you charge borrowers and what you pay for capital. With the 10-year Treasury yield near 4.44%1, use the current rate environment as your starting point for both sides of that spread rather than assuming your cost of capital stays fixed while only borrower rates float.
A forecast that holds cost of capital constant while rates are actually moving will overstate margin in a rising-rate environment and understate it in a falling one, in either case giving you a wrong signal about how much room you actually have on pricing.
Cube for a controller who already models the portfolio in a spreadsheet
If your team already tracks loan-level yield, delinquency, and reserve data in a spreadsheet, even a detailed one, Cube's approach of syncing that spreadsheet against your loan servicing system keeps the calculation logic where it's understood, with less manual export from servicing reports each month.
This matters particularly for a lender whose reserve methodology has real judgment built into it, since that judgment is often easier to preserve and audit inside a spreadsheet formula than inside a dashboard's built-in calculation.
Mosaic for a dashboard tracking portfolio health across vintages
Once your loan book is large enough that tracking yield and delinquency by vintage in a spreadsheet becomes unwieldy, a dashboard consolidating that view can help you spot a developing problem faster. Confirm in a demo that Mosaic can track delinquency and reserve adequacy by loan vintage rather than reporting one blended portfolio figure, since a blended number is exactly what hides an emerging problem in a specific origination cohort.
Also confirm the dashboard can be updated as frequently as your credit risk monitoring actually requires, since a lender managing active credit risk often needs more than a monthly refresh cycle.
Building a stress case into the forecast, not just a base case
Run the portfolio forecast at a base case and a stressed case with elevated delinquency and higher cost of capital, since a lender's business is inherently more exposed to a downturn than most other industries in this comparison series, and a forecast that only shows the base case leaves you unprepared for a credit cycle turn.
Share the stressed case explicitly with your board, lenders, or investors rather than only the base case, since credibility with capital providers depends partly on showing you've actually thought through what happens if conditions deteriorate, not just presented the optimistic scenario.
A working lender forecast covers these points:
- Run the portfolio forecast at a base case and a stressed case with elevated delinquency and higher cost of capital.
- Track interest income, delinquency status, and reserve adequacy together by loan vintage, not as one blended portfolio figure.
- Show delinquent-but-accruing balances separately so a developing credit problem is visible before a formal reclassification.
- Tie the cost of capital assumption to the reference rate your facility floats against and update it as that rate moves.
- Share the stressed case explicitly with your board so a turn in the credit cycle doesn't come as a surprise.
Where Jirav fits a lender scaling origination and servicing staff
Jirav's driver-based approach is useful when you're planning to grow origination volume and need the model to show how many additional underwriters and servicing staff that growth requires, since underwriting and servicing capacity, not just capital availability, is often the real constraint on how fast a specialty lender can responsibly grow its book.
A lender that grows origination volume faster than its underwriting and servicing capacity tends to see credit quality slip first, since overstretched underwriters cut corners under volume pressure, and that erosion in the newest vintages often doesn't show up clearly in the numbers until well after the real damage is already done.
What Good Looks Like
A well-run specialty lender tracks portfolio yield, delinquency, and reserve adequacy by loan vintage, with cost of capital modeled as floating alongside the current rate environment rather than held fixed.
Building The Capability (5-Stage Skill Ladder)
How to Get Started
Disclosure: We may earn a commission if you buy through some links on this page. It doesn't change what we recommend.
Cube fits a lender whose team already models loan-level yield and reserve data in a spreadsheet and mainly wants loan servicing data synced in automatically.
Mosaic is worth a demo once tracking yield and delinquency by vintage in a spreadsheet becomes unwieldy, provided it can report those by vintage rather than as one blended figure.
Jirav suits a lender planning to grow origination volume and wanting underwriting and servicing staffing tied explicitly to that growth rather than assumed to keep pace on its own.
Frequently Asked Questions
Should interest income on a delinquent loan still be recognized as revenue?
Generally yes until the loan is formally classified as non-performing under your accounting policy, at which point accrual typically stops. Track delinquent-but-accruing balances separately in your model so a developing credit issue is visible before it forces a formal reclassification.
How should cost of capital be modeled if it's tied to a floating-rate credit facility?
Tie the model's cost of capital assumption to the same reference rate your facility floats against, updated as that rate moves, rather than a fixed assumption set once at the start of the year. A floating cost of capital that isn't modeled as floating will misstate your margin whenever rates move.
Do Cube or Mosaic calculate credit loss reserves for us?
No, reserve methodology is a judgment call that should follow your accounting policy and be reviewed by your accountant, typically based on historical loss rates by vintage or risk tier. Both tools can incorporate that reserve calculation into a forecast once you've defined the methodology; neither one sets the methodology itself.
Sources
Where we quote a benchmark, we show its source. Other figures in this guide are estimates or general guidance, so check them against your own numbers.
- 10-year US Treasury constant-maturity yield. Federal Reserve H.15 Selected Interest Rates, 2026.
Related Guides
FloQast vs. AuditBoard for Specialty Asset-Based Lenders
Loan loss reserves, nonaccrual interest, and warehouse facility covenants drive a specialty lender's close. See how FloQast and AuditBoard each fit.
Pulley vs. Carta for a Specialty Asset-Based Lender
A worked example of a specialty lender granting equity to its underwriting team, and how Pulley and Carta fit as the balance sheet grows.
409A Valuation for a Specialty Asset-Based Lender
Warehouse covenants can restrict ownership changes, and the loan book itself needs its own valuation before the equity number makes sense. Here's how.
BILL vs Tipalti for Specialty Asset-Based Lenders
Two approaches to AP for specialty asset-based lenders, routine vendor bills versus payee-level structure for co-lending participants.
Sales Tax Exposure Mapping for Asset-Based Lenders
Interest and fee income from lending is exempt almost everywhere. A specialty lender's real sales tax review has to look past the loan book itself.
A Checklist for Whether Pipe or Capchase Fits an Asset-Based Lender
A pitfalls-focused checklist for a specialty asset-based lending firm weighing Pipe or Capchase against warehouse funding for its own working capital.