Cost per delivery and cost per kilometer are two key performance indicators (KPIs) on the same freight spend: Mexican pesos (MXN) of freight plus audited accessorials, divided by deliveries (stops) or by kilometers driven. Same basket, different question: what does each stop cost? what does each lane-km cost?
On Mexico–US distribution the usual failure is comparing base rates only—or mixing units with drop size and km per delivery. This page locks formulas, basket, and an illustrative example labeled for you to calibrate with your operation—not a viral benchmark.
- MXN/delivery · MXN/km
- 2 KPIs
- freight + audited accessorials
- 1 basket
- illustrative, not a target
- Calibrate
- pattern when auditing 100%
- 5–7%
What cost per delivery and per km are
Both are unit-cost KPIs for the distribution leg: from the distribution center (DC) or plant to the customer (or the retailer DC). They do not replace OTIF or % complete shipment file—they measure pesos per unit of work.
Cost per delivery (MXN/delivery) = MXN in the freight basket ÷ number of deliveries (stops with attempt or close-out per your written rule). Cost per km (MXN/km) = the same basket ÷ kilometers in the universe (odometer, TMS, or GPS—pick one source and do not switch mid-quarter).
They exist to compare lanes, carriers, and route designs under the same rule. Without a written rule, the dashboard celebrates one number and accounts payable pays another.
Formula matrix and common mistakes
Freeze the formula before you negotiate. The matrix locks numerator, denominator, and the mistake that usually inflates or deflates the KPI.
Cost per delivery
Operable formula: MXN freight (+audited accessorials) ÷ deliveries
Unit: MXN / delivery
Common mistake: Dividing base rate only; or counting “orders” ≠ stops
Cost per km
Operable formula: MXN freight (+audited accessorials) ÷ km driven
Unit: MXN / km
Common mistake: Using planned km when GPS shows systematic detours
Operable formula: Km driven ÷ deliveries
Unit: km / delivery
Common mistake: Calling it “% efficiency” or mixing with kg
Operable formula: Kg (or pieces) delivered ÷ deliveries
Unit: kg / delivery
Common mistake: Using it as a cost proxy without converting to MXN
Accessorials % of freight
Operable formula: MXN accessorials ÷ MXN base freight
Unit: %
Common mistake: Parking FSC in base one month and in accessorials the next
Basket: what enters the numerator
The basket must match when you compare carrier A vs B or lane X vs Y. If one side includes detention and the other does not, the “cheaper” option is an illusion.
| Line | In the numerator? | Minimum evidence |
|---|---|---|
| Base freight / contracted rate | Yes | Tariff or rate confirmation + trip ID |
| Detention / dwell | Yes (if paid or accrued) | GPS vs free time; typed invoice line |
| Lumper / handling | Yes | Authorization + POD or dock ticket |
| Redelivery / second attempt | Yes | Typed reason + second evidence |
| Fuel surcharge billed separately | Yes, if not in base | Same rule all period |
| Passed-through VAT | Per policy (gross vs net) | Do not mix gross and net on one board |
| Retailer chargebacks / deductions | Not in freight unit cost (separate KPI) | See retailer deduction cluster |
Dock wait playbook: hidden cost of detention. Accessorial definition: what accessorials are.
Unit discipline (do not mix)
Three different metrics get collapsed into one slide. Separate them in writing or the committee will think “efficiency fell” when only drop size changed.
- MXN/delivery and MXN/km — cost (this page).
- Km/delivery — route geometry (km per delivery).
- Kg/delivery — density per stop (drop size).
The distribution KPIs hub corrects publishing “kg/stop” as route efficiency with a 99% target. Here the anchor is pesos—not an invented percentage.
Illustrative MXN example (calibrate)
Illustrative numbers for one regional delivery day. Not a client benchmark or published target. Calibrate with your tariff, odometer, and stop rules.
| Data (illustrative) | Value | KPI |
|---|---|---|
| Deliveries (stops) | 40 | MXN/delivery denominator |
| Km driven | 480 km | MXN/km denominator |
| Audited base freight | $24,000 MXN | Part of numerator |
| Audited accessorials (detention + lumpers) | $4,800 MXN | Part of numerator |
| Total basket | $28,800 MXN | Shared numerator |
| Cost per km | $28,800 ÷ 480 = $60 MXN/km | Distance unit cost |
| Cost per delivery | $28,800 ÷ 40 = $720 MXN/delivery | Stop unit cost |
| Km per delivery (context) | 480 ÷ 40 = 12 km/delivery | Not a cost |
When to use MXN/delivery vs MXN/km
Do not pick the KPI that “looks better.” Pick the one that triggers Monday’s decision.
| Decision | Primary KPI | Support KPI |
|---|---|---|
| Raise drop size / fewer visits? | MXN/delivery + drop size | OTIF, chargebacks |
| Is the route too capillary? | Km/delivery + MXN/delivery | MXN/km, empty % |
| Carrier A vs B on the same lane? | Audited MXN/km | OTIF, % complete file |
| Cost-to-serve by customer? | MXN/delivery by customer | Frequency, drop size |
| Release invoice payment? | % complete file + audited lines | Typed accessorial MXN |
How it works on Mexico–US lanes
On domestic Mexico or USMCA truck freight, unit cost lives with DC appointments, free time, crossing or transfer, CFDI with Carta Porte, and POD. A “pretty” MXN/km without a complete file does not help accounts payable.
Service context (not cost): the 4th National Logistics Indicators Study 2026 (#SoyLogístico / LDM / EGADE) reports customer delivery fill rate near ~93% average in its sample—useful so you do not confuse low cost with acceptable service. Calibrate cost targets to your lane (e.g. Monterrey–Laredo vs metro capillary), not a national average.
Cost-to-serve depth: distribution costs in Mexico.
How to instrument the KPI
The number lives in the cross of trip, km, stops, and paid lines—not in a forever spreadsheet disconnected from GPS.
Instrumentation
From invoice to unit cost
Freeze basket
Base + accessorials
Lock km source
GPS or TMS
Count deliveries
Written rule
Audit lines
Before pay
Publish dual
MXN/km and /delivery
Checklist for a defensible unit cost
If you cannot check the first four items on one lane, you do not need a twenty-series dashboard yet—you need a rule and evidence.
Elige un paso para ver el detalle
Detalle del paso · 01
Write the basket
Step 1
OCL and 100% audit
OCL Cargo is an autonomous transportation management system (TMS) with agents (computer use: the agent operates screens). It does not sell “another MXN/km chart” as the thesis: it runs the pre-pay cross of rate, CFDI, Carta Porte, GPS, and POD so the unit-cost numerator is what you actually paid.
When you leave sampling and audit 100% of the pilot flow, the typical pattern is recovering 5–7% of freight spend. That recovery usually moves MXN/km and MXN/delivery more than a cosmetic route redesign that never cleans accessorials. It coexists with your system of record; your team decides exceptions.
Related reading
Key takeaways5 points
- Cost per delivery and cost per km are the same freight spend with two denominators: stops vs kilometers.
- Numerator = MXN freight + audited accessorials; without detention and lumpers, the KPI understates cost.
- Do not mix with drop size (kg/stop) or an invented “% efficiency”: each KPI keeps its unit.
- Illustrative example (calibrate): $28,800 MXN ÷ 480 km = $60/km; ÷ 40 stops = $720/delivery.
- When auditing 100% of the pilot, the typical OCL pattern is recovering 5–7% of spend — that moves true unit cost.
Does your MXN/km ignore the detention you still pay?
Frequently asked questions
They are two unit-cost key performance indicators (KPIs) on the same freight spend. Cost per delivery = Mexican pesos (MXN) of freight (+ audited accessorials) ÷ number of deliveries (stops). Cost per km = those same MXN ÷ kilometers driven. Same basket, different denominator.
Yes, if you want a defensible cost. The minimum basket is base freight + audited accessorials (detention, lumpers, redelivery, fuel when billed separately). Comparing base rates alone hides the leak.
MXN/delivery when you decide stop density, drop size, or cost-to-serve by customer. MXN/km when you compare lanes or carriers on distance. Capillary routes usually need MXN/delivery; long linear lanes often need MXN/km.
Km per delivery is distance ÷ stops (route efficiency). Drop size is kg (or pieces) ÷ stops. Unit cost turns that geometry into pesos. Do not publish “93% efficiency” by mixing kg with km.
Do not treat a viral number as a blind target. The 2026 National Logistics Indicators Study (#SoyLogístico / LDM / EGADE) gives service context (e.g. ~93% average delivery fill rate in its sample), not a universal unit cost. Calibrate with your tariff and lane.
It crosses rate, CFDI (digital tax invoice), Carta Porte, GPS, and proof of delivery (POD) before pay. When auditing 100% of the pilot flow, the typical pattern is recovering 5–7% of freight spend — invented or mistyped accessorials are often in that basket.