Amazon DSP Profit Margins: What Owners Actually Keep

How the revenue actually arrives

A DSP's revenue is a weekly settlement built from contracted components: a fixed monthly amount per route commitment, per-route rates, and per-package or per-stop components, plus the piece that swings everything, scorecard-linked bonuses tied to safety, delivery quality, and reliability metrics. The settlement lands as one deposit; inside it are the line items that tell you which routes and which weeks made money. Booking the deposit as 'revenue' and moving on, which is what generic bookkeeping does, erases exactly the information a DSP runs on.

The cost structure, honestly

• Driver labor: 55–65% of revenue, the dominant line. Amazon requires W-2 drivers (no 1099 shortcut exists here), which means full payroll burden, taxes, workers' comp, and the overtime that route overruns quietly generate
• Vehicle costs: leases through the program's providers, insurance, fuel where applicable, and, the sleeper, damage: chargebacks, repairs, and end-of-lease reconciliation on vans that live a hard life
• Turnover: the invisible line. Driver churn in last-mile routinely exceeds 100% annually; every departure costs recruiting, onboarding hours, training pay, and route-quality dips that hit the scorecard. DSPs that halve their turnover change their P&L more than any negotiation with Amazon ever will
• Overhead: dispatcher(s), your ops manager, phones, uniforms, office, modest individually, meaningful in sum on thin margins

Why the scorecard IS the margin

Run the math on a typical mid-size DSP: 30 routes, roughly $2.5M in annual revenue. Base economics after labor, vehicles, and overhead might leave 5–8%: $125K–$200K. The scorecard bonus tiers ('Fantastic Plus' and friends) can swing several percent of revenue. On these numbers, the difference between top-tier and mediocre scorecard performance is the difference between the profit Amazon's brochure promised and a breakeven year. That's why the operational metrics and the financial statements have to be read together, a DSP owner watching the P&L but not the per-route scorecard economics is watching the scoreboard of last month's game.

What profitable DSPs' books look like

• Settlements reconciled line-by-line, weekly, every component verified against the contract; shortfalls disputed while they're fresh
• Per-route P&L: revenue components, driver hours (with OT), and vehicle costs landed on each route, revealing the two routes that quietly lose money in every fleet
• Labor percentage tracked weekly, not discovered quarterly, with OT flagged the week it spikes, because route overruns compound
• Damage and chargeback tracking by van and by driver, patterns are coachable; totals are just losses
• Turnover costed, when replacement cost per driver is a known number, retention spending stops looking like an expense

Common DSP bookkeeping failures

The pattern we see in DSP files that arrive for cleanup: settlements booked net as single deposits (making revenue analysis impossible), payroll in one lump with no route allocation, vehicle chargebacks buried in 'miscellaneous,' and no accrual for the end-of-lease van reconciliation that everyone knows is coming. Every one of these is fixable, and fixing them is usually worth more than any operational change, because you can't manage a margin you can't see. Courier & Last-Mile Delivery ↗