How to Build an LTL Carrier Scorecard
Build an LTL carrier scorecard with eligibility gates, five defined KPIs, fair comparison groups, sample-size rules, and a practical review process.
An LTL carrier scorecard turns shipment records into a repeatable carrier-selection tool. Start with eligibility, then measure each carrier on comparable lanes using a small set of clearly defined service, billing, freight-condition, and response metrics. Keep the raw counts beside every percentage so a perfect result from two shipments does not outrank a solid result from fifty.
The goal is not to crown one carrier as universally best. It is to identify which eligible carrier has performed well for the shipment profile you are about to quote.
Start with an eligibility gate, not a score
Before performance points are assigned, confirm that the carrier is eligible for the work. For interstate carriers, record the legal name and USDOT or docket number, then check current operating-authority and insurance information in the Federal Motor Carrier Safety Administration’s Licensing and Insurance system. Review the carrier’s FMCSA SAFER Company Snapshot separately. SAFER can show identification, operation type, cargo information, inspection and crash summaries, and a safety rating when one exists.
Treat those government records as screening inputs, not as a homemade safety grade or a substitute for the review your organization requires. Save the verification date because public records can change.
An eligibility worksheet can be simple:
| Gate | Evidence to retain | Result |
|---|---|---|
| Identity | Legal name, USDOT number, and docket number when applicable | Pass, fail, or review |
| Operating authority | Current authority record for the planned service | Pass, fail, or review |
| Insurance filing | Current FMCSA filing information relevant to the carrier | Pass, fail, or review |
| Safety review | Dated SAFER snapshot and any required internal review | Pass, fail, or review |
| Service fit | Written confirmation that the carrier handles the lane, commodity, equipment, and accessorial needs | Pass, fail, or review |
A failed gate should not be averaged away by a high service score. Escalate the record for review or exclude the carrier from that shipment’s comparison.
Define the comparison group before choosing metrics
Score like against like. At minimum, separate results by the dimensions that materially change the work:
- origin and destination region or lane;
- service type and delivery requirement;
- commodity and packaging profile;
- freight class or density profile when relevant;
- accessorial requirements such as liftgate, limited-access, residential, appointment, or inside service; and
- reporting period.
This matters because LTL rating and handling are sensitive to the freight presented. The National Motor Freight Traffic Association explains that NMFC freight class reflects density, handling, stowability, and liability. A carrier’s results for compact, easy-to-handle freight should not automatically predict performance for long, fragile, non-stackable, or otherwise difficult freight.
Use five metrics with explicit formulas
Choose definitions before looking at the results. The following structure is a practical starting point, not an industry-mandated benchmark.
| Metric | Formula | What to record |
|---|---|---|
| Pickup reliability | Qualifying on-time pickups divided by pickups with a documented pickup window | Requested window, accepted window, actual pickup time, exclusion reason |
| Delivery reliability | Qualifying on-time deliveries divided by deliveries with a documented delivery standard | Standard used, actual delivery time, appointment or exception notes |
| First-pass invoice accuracy | Invoices matching the accepted rate scope without an unexplained correction divided by audited invoices | Quoted scope, invoice total, adjustment reason, responsible party |
| Freight-condition exception rate | Shipments with a documented shortage or damage exception divided by delivered shipments | Delivery notation, photos when available, claim reference, packaging notes |
| Exception response | Exceptions receiving a substantive response within your written target divided by exceptions requiring carrier action | Opened time, first substantive response, owner, resolution date |
Define “qualifying” and “substantive” in the scorecard instructions. For example, a status acknowledgment is not necessarily a substantive response, and a missed pickup caused by freight not being ready should not automatically count as a carrier failure. Preserve the excluded shipment count and reason; silently removing unfavorable records makes the score unreliable.
Cost belongs beside these metrics, but it needs its own disciplined view. Compare the accepted quote, final invoice, and reason-coded differences. A higher final invoice is not automatically a billing error. Current carrier rules can allow charges or corrections for circumstances such as reweighs, reclassification, dimensional conditions, accessorial service, storage, or redelivery. Check the tariff or written pricing agreement that governed the shipment, and separate shipper-data errors from carrier billing errors.
Weight only what the team can measure consistently
Weights should reflect the business consequence of each metric and total 100%. One illustrative model is:
| Metric | Illustrative weight |
|---|---|
| Delivery reliability | 30% |
| First-pass invoice accuracy | 25% |
| Freight-condition exception rate | 20% |
| Pickup reliability | 15% |
| Exception response | 10% |
Convert each metric to a 0-to-100 score using a documented rule, multiply by its weight, and add the weighted values. Do not copy these weights blindly. A project-delivery operation may emphasize delivery reliability, while a fragile-freight program may place more weight on condition and claims handling.
Always display the denominator, reporting period, and missing-data treatment next to the weighted score. If a carrier has too few comparable shipments, show “insufficient data” rather than filling the gap with a neutral score. A weighted total without exposure counts creates false precision.
Work through a sample without inventing a benchmark
Suppose two eligible carriers served the same lane and service profile during the review period:
| Carrier | Comparable delivered shipments | On-time deliveries | Audited invoices | Accurate on first review | Condition exceptions |
|---|---|---|---|---|---|
| Carrier A | 18 | 16 | 18 | 17 | 1 |
| Carrier B | 7 | 7 | 7 | 6 | 0 |
Carrier B has a perfect delivery result in this small sample, while Carrier A has more observations and a stronger invoice result. The right conclusion is not that one carrier is automatically superior. Show the calculated rates, flag Carrier B’s smaller sample, review the lane and shipment mix, and consider whether the next shipment resembles the records behind either result.
This example is illustrative only. Set minimum sample rules from your own shipment frequency and decision risk. When volume is sparse, use the scorecard as structured evidence alongside current quote terms, service fit, and documented exceptions rather than forcing a definitive rank.
Keep claims and damage data in context
Track at least three distinct facts: delivery exceptions, claims filed, and claims resolved. They are not interchangeable. A delivery notation may not become a claim, and an open claim should not be treated as a completed carrier outcome.
Also retain the governing documents. NMFTA’s Uniform Straight Bill of Lading terms address carrier liability, exceptions, and claim-filing requirements, while allowing a written agreement or carrier tariff to establish different or additional claim provisions. For each event, record the delivery date, notation, filing date, documents supplied, current status, resolution date, and outcome. Confirm the terms that actually applied instead of treating a general timeline as universal.
Review causes before changing the routing guide
A useful monthly or quarterly review asks four questions:
- Did each metric use the same definition across carriers?
- Were excluded records documented and applied consistently?
- Did a lane, commodity, packaging pattern, or accessorial requirement drive the result?
- Does the issue belong to the carrier, the shipper’s data, the consignee, or an unresolved handoff?
Discuss the underlying records with the carrier, assign corrective actions, and record an owner and review date. Keep current-period and trailing-period views side by side so one unusual shipment does not erase a longer pattern. When definitions or weights change, version the scorecard rather than rewriting history.
Use the scorecard during rate comparison
Apply the scorecard only after confirming the new shipment fits the comparison group. Prepare the origin and destination ZIP codes, packaged dimensions and weight, commodity description, freight class when known, ready date, delivery requirements, and needed accessorial services. Then compare the current rate and service terms with the carrier’s relevant lane-level record.
Shipocity is backed by a team with more than 40 years of combined logistics experience. Through established industry relationships, the platform helps businesses compare competitive freight rates for their specific shipment.
Compare rates for your next shipment
