Cube vs. Mosaic for Freight Margin and Fleet Cost Modeling

Two loads with the same revenue can have very different margins once you factor in deadhead miles to reposition the truck, fuel surcharge recovery on that specific lane, and whether the driver was a company employee or an owner-operator paid on a different settlement structure. Averaging all of that into one fleet-wide margin figure tells you the fleet is profitable while individual lanes quietly lose money.

The fix is a model that calculates margin at the load or lane level and rolls up from there, not a top-down blended number that gets adjusted after the fact. A lane that looks fine on the fleet average can be the one dragging every other lane's margin down once you break it out.

Many independent trucking operations and smaller 3PLs still build per-load profitability in Excel because dispatch and fuel card systems rarely export the exact margin view an owner wants. Cube's approach, syncing that existing spreadsheet against your dispatch, fuel, and payroll data, fits when the per-load logic is already correct and the pain point is manual data entry, not the math itself.

This is the stronger starting point if whoever owns the model today is confident in it and just wants the weekly settlement pull automated, rather than rebuilding the entire margin calculation inside a new tool.

Once a fleet grows past a size where one person can eyeball every settlement, a dashboard that surfaces margin by lane, driver, and truck without anyone rebuilding a pivot table each week becomes genuinely useful. Confirm during a demo that Mosaic can handle a driver-pay structure that mixes company drivers and owner-operators, since that split materially changes how cost per mile should be calculated, and a generic template can flatten a distinction that matters to your margin.

Fuel and driver pay are the two variables most likely to move your margin more than a change in freight rates will, so build the forecast with both as adjustable drivers rather than fixed shares of revenue. A common mistake is forecasting next quarter's margin off this quarter's fuel price without stress-testing what happens if diesel prices move meaningfully in either direction, since that swing hits every loaded and deadhead mile in the fleet at once.

Run the forecast at a baseline, a higher-fuel-cost case, and a lower-fuel-cost case so a rate swing doesn't blow up your cash plan without warning. A fleet that only ever models the baseline case tends to be the one caught off guard when fuel moves fast.

Jirav is worth considering when you're modeling growth in truck count and driver headcount together, since adding a truck without a driver lined up, or the reverse, is a common way fleets overspend on equipment relative to what they can actually run. A driver-based forecast that ties truck purchases, driver hiring, and maintenance reserve to the same growth assumption keeps those three from drifting apart.

A truck's maintenance cost isn't flat across its life; it climbs as mileage accumulates, and a fleet growing its average truck age will see maintenance eat into margin even if every other input stays the same. Build a per-truck maintenance reserve into the forecast tied to mileage or age bands rather than a flat per-truck average, so an aging fleet's rising cost shows up in the plan before it shows up as a surprise repair bill.

A fleet that replaces trucks on a fixed schedule regardless of actual condition avoids some of this risk but pays for it in higher fixed equipment cost; one that runs trucks longer needs the maintenance reserve to be more precise, since the range of outcomes on an older truck is wider than on a newer one.

Before committing to either platform, export a month of settlements and dispatch data and see how cleanly it maps into the tool's expected format; a mismatch here is the most common reason a fleet's first month on a new system produces numbers nobody trusts. Ask specifically how the tool treats a driver who ran under two different pay structures within the same month, since that's a real scenario for a fleet transitioning drivers between company and owner-operator status.

Test these points before moving your margin model:

• Export a month of settlements and dispatch data and see how cleanly it maps into the format the tool expects.
• Ask how the tool treats a driver who ran under two different pay structures within the same month.
• Confirm it can handle a mix of company drivers and owner-operators, since that split materially changes cost per mile.
• Verify that deadhead miles and fuel surcharge recovery can be calculated lane by lane instead of blended across the whole fleet.
• Make sure fuel price and driver pay can each be adjusted as separate drivers in the forecast.