How Do You Measure Carrier Performance?

Carrier performance does not manage itself. Left unmonitored, underperformance becomes the baseline, rate increases go unchallenged, and the operational cost of a carrier that is not delivering shows up everywhere except in a report that would make it actionable. Measuring carrier performance correctly is what gives operations teams the data to hold carriers accountable and make better decisions about where volume should go.

Start With the Right Metrics

Measuring carrier performance starts with choosing metrics that reflect operational reality rather than what is easiest to pull from a carrier portal.

On-time delivery rate is the primary metric. It measures the percentage of shipments delivered by the promised date, calculated at the carrier and service level, not in aggregate across all providers. Aggregate on-time rates are useful for headline reporting. Carrier-specific rates are useful for decisions.

Transit time variance measures consistency rather than just compliance. A carrier that sometimes delivers in two days and sometimes in five, with an average that meets the service window, is an unreliable carrier regardless of whether the average looks acceptable. Measuring the standard deviation of transit times by lane surfaces consistency problems that on-time rates can miss.

Exception rate tracks the frequency of carrier-reported delivery issues including attempted deliveries, address problems, weather holds, and lost packages, per carrier and lane. A rising exception rate on a specific carrier or lane is an early signal worth catching before it compounds into a customer experience problem.

Damage and loss rate measures claims filed as a percentage of total shipments per carrier. This does not appear on a carrier invoice but it appears consistently in the P&L through reshipment costs, refunds, and replacement inventory. Carriers with high damage rates are more expensive than their rate cards suggest.

Claims resolution rate tracks how effectively a carrier handles claims when things go wrong. A carrier with a high damage rate and a slow, difficult claims process is a significantly worse operational partner than one whose rate card looks identical but resolves claims quickly and fairly.

Pull Data at the Right Level of Granularity

Carrier performance measurement only produces actionable insight when the data is segmented correctly. Measuring at too high a level hides the variation that matters.

Break on-time delivery rate down by carrier, service level, origin location, destination zone, and time period. A carrier averaging 94% on-time overall may be running at 88% on-time into zones six through eight during Q4. That is a specific, addressable problem. The aggregate number does not reveal it.

Seasonal segmentation matters too. A carrier that performs well through Q1 to Q3 but degrades during peak season represents a different operational risk than one with consistent year-round underperformance. Understanding the seasonal pattern is what allows brands to adjust carrier mix proactively before peak rather than reactively during it.

Establish a Baseline Before Drawing Conclusions

A single week of data is not carrier performance measurement. It is a data point. Meaningful carrier performance assessment requires a baseline established over a sufficient time period to distinguish genuine performance trends from normal variability.

Establish a rolling baseline for each carrier by service level and lane. Three months of data provides a reasonable foundation. From there, weekly and monthly performance can be compared against the baseline to identify genuine degradation rather than normal fluctuation.

When performance moves meaningfully below the baseline and stays there, that is a signal worth acting on. When it dips briefly and recovers, that is variability rather than a structural performance issue.

Connect Measurement to Decisions

Carrier performance data has no operational value sitting in a report. Its value is in the decisions it drives.

Routing adjustments should happen automatically when carrier performance on a specific lane drops below a defined threshold. The operations team should not have to manually review performance data and decide to reroute. The routing logic should do it.

Carrier conversations should be grounded in specific data rather than general impressions. A carrier underperforming against contracted service levels on documented lanes is a different conversation than a general complaint about reliability. Specific data produces specific responses.

Volume allocation decisions should reflect current performance, not historical relationships. Carriers that perform well earn volume. Carriers that underperform lose it. That dynamic only exists when performance is being measured consistently and the data is feeding into routing decisions.

Want to build a carrier performance measurement framework that actually drives decisions? Talk to one of our shipping experts. Book a demo.