The Real Cost of Last-Mile Inefficiency for Logistics Providers: Failure Rate Moves Margin Faster Than Cost

The same failure, two different balance sheets

Take an operation running 10,000 stops a day with a 3% first-attempt failure rate. Three hundred failures.

If those stops belong to a retailer delivering with its own fleet, the 300 failures are 300 units of additional cost, absorbed into a cost-to-serve line that the retailer can respond to. It can widen a delivery window, raise a free-shipping threshold, change the promise, or price the channel differently. The failure is expensive and the business retains its pricing levers.

If those stops belong to a logistics provider delivering under a per-delivery contract, the same 300 failures are something else. The provider performed 300 deliveries that generated no billable event, will perform 300 more attempts at full cost, and may owe a service credit on top. Revenue for the period was set when the contract was signed. Cost for the period just rose.

That is the distinction this piece is about, and it is not a rhetorical one. For a shipper, inefficiency is a cost problem with pricing remedies. For a provider, inefficiency is a margin problem with only operational remedies, because the price is fixed until renewal.

The existing published work on failed delivery cost is largely written from the shipper’s side, and correctly so: shippers carry customer service, compensation, and returns costs that providers usually do not. The provider’s exposure is narrower in scope and considerably sharper in effect, and it deserves its own arithmetic.

Also Read: Failed Delivery Cost Framework: The Hidden Cost Categories of Failed First Attempts in U.S. Last-Mile Operations

Why the provider’s exposure inverts the shipper’s

Four structural differences, and each changes which lever matters.

Revenue is fixed at contract, cost is variable at execution. A shipper’s failure raises its own cost and it can adjust price or promise in response. A provider’s failure raises cost against revenue that cannot move, so the entire delta lands in margin rather than being shared with a customer.

The penalty flows to the provider. Service credits and SLA penalties are contractual obligations of the party performing the service. For a shipper, a missed window is a customer experience problem. For a provider it is also a liability with a number attached.

Repricing is unavailable until renewal. This is the constraint that shapes everything else. A shipper facing a rising failure rate has commercial options available this quarter. A provider has execution, and only execution, until the contract comes up.

Margin per unit is thin, so the denominator is small. This is where the arithmetic gets uncomfortable, and it is worth doing explicitly.

The margin arithmetic

Three inputs, all of which a provider already knows. Use your own numbers; the values below are illustrative and chosen to show the shape rather than to serve as benchmarks.

Let R be revenue per successful delivery, C be direct cost per attempt, and S be first-attempt success rate.

Where a failed attempt is retried once and then succeeds, cost per successful delivery is approximately C × (2 ? S), because every hundred orders consume a hundred first attempts plus one attempt for each failure.

Take R = $6.00 and C = $4.80, which is an 80% cost ratio, and a first-attempt success rate of 97%.

• Cost per success: 4.80 × 1.03 = $4.94
• Margin per success: 6.00 ? 4.94 = $1.06

Now let first-attempt success fall three points to 94%, which is the kind of drift a peak season, a new territory, or a cohort of new drivers can produce without anyone declaring a crisis.

• Cost per success: 4.80 × 1.06 = $5.09
• Margin per success: 6.00 ? 5.09 = $0.91

Cost per success rose by under 3%. Margin fell by roughly 14%.

That ratio is the whole argument. Because margin is the small difference between two larger numbers, a change in failure rate is amplified into margin by roughly five times in this example. A provider looking at a cost line will see a rounding error. The same movement in the margin line is a material deterioration, and it is the line the business is actually judged on.

Run the same calculation at your own cost ratio. The thinner the margin, the higher the amplification, which means the providers least able to absorb a failure-rate drift are the ones most exposed to it.

One note on external benchmarks. The commonly cited figure of roughly $17.78 per failed delivery is a shipper-side total that includes support handling, compensation, and returns flow. A provider’s direct reattempt cost is smaller than that. Its margin exposure, as the arithmetic above shows, is proportionally larger. Using the shipper figure as a provider input overstates the cost and understates the damage.

Also Read: Fleet Utilization for 3PLs: How to Maximize Asset Performance Across a Multi-Client, Multi-Fleet Operation

Five cost lines a provider carries that a shipper does not

The unbillable attempt. Work performed, fuel burned, driver hour consumed, no billable event generated. This is the line most provider cost models capture, and it is the smallest of the five.

SLA credit exposure. Contractual, quantified, and payable. Worth tracking as a percentage of contract revenue rather than as an absolute, because that is how it will be discussed at renewal.

Dwell you do not control. Time waiting at a shipper’s dock or a customer’s site is provider cost incurred to serve someone else’s process. ATRI found drivers were detained at 39.3% of all stops in 2023, losing between 117 and 209 hours per year depending on sector. Where that time is not billable as an accessorial, it is margin transferred to the counterparty.

Capacity displacement. The subtle one, and usually the largest. In a capacity-constrained operation, a reattempt does not cost the reattempt. It costs the billable delivery that could have occupied the same slot. If your network is running full, the true cost of a failed delivery is the marginal revenue you could not serve, which is a considerably bigger number than the direct cost and appears in no report.

Renewal exposure. Today’s performance data is an input to tomorrow’s pricing negotiation. A failure rate that costs margin this year also weakens the position from which you argue for a rate increase next year, which makes it the only cost line on this list that compounds.

The fourth line is worth sitting with. Most provider cost analysis treats a reattempt as an incremental cost against spare capacity. That assumption holds in a slack network and fails precisely when it matters, which is during peak, when capacity is tight and failure rates are highest at the same time.

Also Read: Last-Mile Delivery in 2026: Costs, Challenges, Fixes

Shipper exposure and provider exposure compared

Dimension	Shipper	Logistics provider
What a failed delivery is	An added cost	Unbillable work against fixed revenue
Where it lands	Cost to serve	Margin
Amplification	Roughly proportional to cost	Multiplied by the inverse of margin ratio
SLA consequence	Customer dissatisfaction	Contractual credit or penalty
Available levers	Price, promise, threshold, channel	Execution only, until renewal
Support and returns cost	Carried	Usually not carried
Largest hidden line	Retention effect	Capacity displacement
Compounding effect	Brand erosion	Weakened renewal position

The row that decides strategy is the fifth. A shipper can respond commercially this quarter. A provider’s only available response is to raise first-attempt success and to recover capacity, which is why operational precision is a margin instrument for a provider rather than a service refinement.

Also Read: How Route Optimization Cuts Last-Mile Delivery Costs 2026

The levers a provider actually has

Four, in descending order of effect.

First-attempt success. The direct lever, and the one with the amplification behind it. Its drivers are unglamorous: address quality and geocoding accuracy, whether the promised window matched what the network could serve, access information, and whether the recipient was reachable. Most provider failure analysis stops at reason codes and never reaches the plan that produced them.

Capacity recovery on reattempts. Whether a failed delivery is retried on a dedicated run or absorbed into an existing route. The difference is most of the reattempt cost, and it depends on whether reattempts are planned alongside forward volume rather than queued separately.

Evidence for penalty defence. For a provider, proof of delivery quality and reason-code specificity are commercial controls rather than service niceties. A penalty is defensible only where the record shows the failure was customer-caused or shipper-caused, with time, location, and attempt evidence attached. Coarse reason codes mean every disputed failure is conceded by default.

Dwell measurement and accessorial recovery. Time waiting is either billable or it is a gift. Recording it precisely is the precondition for charging for it, and for arguing at renewal that the shipper’s site behaviour is part of the cost base.

Note what is absent from this list: price. That is the point.

Also Read: Predictive Capacity Planning: The Peak Season Business Case

What to measure

Margin per successful delivery, not cost per attempt. The single most important change. Cost per attempt understates the effect of failure by roughly the amplification factor calculated above.

Cost per success against cost per attempt. The gap between them is your failure tax, expressed in the same unit as your pricing.

Unbillable attempt rate. Attempts performed that generated no billable event, as a share of total attempts. This is the cleanest single measure of operational waste in a per-delivery contract.

SLA credit exposure as a percentage of contract revenue, by client. Reported monthly, before it becomes a renewal conversation.

Attributable failure share. The proportion of failures with evidence sufficient to attribute cause to the customer or the shipper rather than to you. This is your penalty defence position, measured.

Displaced billable capacity. During constrained periods, the billable deliveries not served because capacity was consumed by reattempts. Hardest of the six to produce and the one that changes the size of the problem.

How Locus improves provider margin rather than shipper cost

Locus, the world’s first Decision-Intelligent, Agentic TMS, is built for the execution levers above, which is the relevant half for a provider who cannot reprice. Within its DiSCO framework, the Digital Supply Chain Officer, specialized agents run a continuous Sense-Decide-Execute-Learn cycle against a model of more than 250 real-world constraints.

Four capabilities map to the four levers.

First-attempt success. Geocoding resolves ambiguous and non-standard addresses to precise coordinates, and access attributes, time windows, and recipient constraints participate in the plan rather than sitting beside it, so the promise made is one the network can serve.

Reattempt on existing capacity. Because forward and recovery work are planned in the same environment against the same constraint set, a failed delivery can be reinserted against live capacity across the fleet rather than queued for a dedicated run, which is where most of the reattempt cost sits.

Evidence at the doorstep. Proof of delivery is captured in the driver application with timestamp and location, and failure reasons are recorded as structured attributes rather than free text, which is what makes an attribution defensible when a client disputes a credit.

Dwell as measured state. The Hub agent models facility readiness and dwell, which turns waiting time from an absorbed cost into a recorded one, and a recorded one can be charged or renegotiated.

The Settlement agent closes the commercial loop, running invoice creation, reconciliation, and payment release as one workflow, so billable events and performed work are reconciled rather than diverging.

Locus has processed more than 1.5 billion deliveries for 360-plus enterprise customers across 30-plus countries at 99.99% uptime, orchestrating over 1,000 carriers, with more than $320 million in aggregate logistics cost savings. It is recognized by Gartner for seven consecutive years, featured in the 2026 Hype Cycle for Supply Chain Execution and Logistics Technologies, named a Leader in TMS by QKS Group (SPARK Matrix), and ranked #1 in Route Planning on G2’s 2026 Best Software Awards. In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Locus continues to operate independently. Further analyst recognition is published in full.

Two provider-side deployments show the two halves of the argument.

A Fortune 50 parcel and logistics provider centralized dispatch across 51 sites in a 120-country network, running more than a million freight shipments a year against a 4,500-strong driver pool split between captive and third-party capacity. Weekly execution rate rose from 75% to 92%. For a provider, execution rate is billable completion, so a 17-point improvement is revenue conversion rather than a service statistic. The deployment also surfaced more than $14 million in contracted capacity that had never been used, including $565,000 at a single site, which is the capacity displacement line measured from the other direction: capacity paid for and not converted into billable work.

A global field service operation across more than 25 US states scheduled appointment-based work against technician skills, per-jurisdiction contracts, differing labor rules, and SLA commitments. SLA penalty risk fell 20%, alongside 18% lower fuel spend and 15% less drive distance and time. The SLA figure is the provider-specific one: a penalty avoided is margin retained on revenue already contracted, which is exactly the line a shipper-side cost model does not contain.

Model your own exposure with the Locus ROI calculator, or request a provider margin assessment to calculate your cost per success against cost per attempt and size your displaced billable capacity during constrained periods.

Recalculate one number this week

Take last month, one contract, and three figures you already hold: revenue per successful delivery, direct cost per attempt, and first-attempt success rate.

Compute cost per success as cost per attempt multiplied by two minus your success rate. Subtract it from revenue. That is your real margin per delivery, and it is lower than the number currently in your management pack if that number was built on cost per attempt.

Then recompute it with first-attempt success one point lower and one point higher. The spread between those two results is what a single point of execution is worth to you, per delivery, every day.

Most providers discover that one point of first-attempt success is worth more than the rate increase they have been preparing to ask for, and unlike the rate increase, it does not require the client to agree.