Carrier scorecards: a score is only useful if it changes the next tender

Key Takeaways

A carrier scorecard should change who gets the next tender. If it changes nothing, it is a report, not a tool.
Use performance in two steps: a gate that decides eligibility, then a ranking among eligible carriers.
A high blended score can hide the one failure a specific load cannot tolerate.
The best performance evidence comes from the shipment records of loads actually run.
Wilson, by Cartage, does not keep a standing carrier scorecard by default.
Wilson can produce carrier-performance reports on request, from the loads it runs.
The company uses that evidence to update its approved carriers and tender rules. Wilson then tenders within those rules.

How should a carrier scorecard influence the next tender?

A carrier scorecard should influence the next tender by deciding who is eligible first, and who ranks highest second. Most scorecards skip the first step. They blend everything into one number and sort by it.

Here is how that goes wrong. Carrier X scores 92 out of 100. It is fast to quote, cheap, and invoices cleanly. But it has missed three of its last ten pickups on refrigerated loads. Carrier Y scores 85. It is slower to quote but has not missed a reefer pickup all quarter. A blended score sends the next reefer load to Carrier X.

The problem is not the score. It is how the score is used. Missed pickups on a reefer load are not a small deduction. They are a reason to stop tendering that load type to that carrier.

So separate the two questions. First, can this carrier take this load? Second, among the carriers that can, who should get it first? A scorecard that answers both in one number answers neither well.

Gate first, rank second: a two-step framework

The two-step framework uses hard rules to remove carriers, then soft measures to order the rest. Each step uses different evidence.

Step one is the gate. A gate is a pass or fail rule. If a carrier fails, it does not get this tender, whatever its other results. Typical gates include:

The carrier is on the company's approved list.
It runs the equipment the load needs.
It serves the lane and the service level.
Its recent missed-pickup count on this load type is below the company's threshold.

Step two is the ranking. Ranking orders the carriers that passed the gate. Typical ranking measures include rate against target, on-time delivery, quote response time, and invoice accuracy.

The gate protects the load. The ranking protects the budget and the service level. Mixing them lets a cheap carrier buy its way past a reliability problem. For a structure that applies these rules to selection, see Approved carrier selection rules: a framework.

Which measures belong in a carrier scorecard

A useful carrier scorecard uses five measures, each taken from a record the company already has. A measure without a record behind it is an opinion. The table lists each measure, its source, and its role in the next tender.

Measure	Where the evidence comes from	Gate or rank?	How it changes the next tender
Tender acceptance	Tender sent and carrier reply on each load	Rank	Carriers that decline often move lower in the order
On-time pickup	Confirmed window vs. actual pickup event	Gate when severe, rank otherwise	Repeated misses on a load type remove the carrier for that type
On-time delivery	Delivery window vs. proof of delivery	Rank	Late carriers lose priority on time-sensitive lanes
Quote response time	Quote request time vs. reply time	Rank	Slow quoters get fewer urgent loads
Invoice accuracy	Booked rate vs. invoiced amount	Rank	Frequent billing gaps lower the carrier's priority

Two rules keep these measures fair. Compare carriers on the same lanes and modes, not across the whole network. And count enough loads to mean something. A carrier with two loads and one miss is not a 50% failure. It is two loads.

Quote response time is the measure most teams lack. It rarely sits in any system. For how to capture it, see How to measure carrier response time.

When the score should not decide the tender

Four situations call for a person's judgment over the scorecard, even when the numbers are clear. These are exceptions, not loopholes, and each one should be recorded.

A new carrier with little history. A short record says little. The company may want to give it test loads on purpose.
A one-off disruption. A carrier that missed pickups during a regional storm may not have a lasting problem.
A contract commitment. The company may owe a carrier a share of volume under a signed agreement.
A customer-specified carrier. Some customers route their own freight. The scorecard does not override that.

In each case, someone should write down why the tender went against the ranking. That note matters later. Without it, the next person sees a low-ranked carrier winning loads and cannot tell why.

Price is another trap. The cheapest carrier is not always the best tender, even when it passes the gate. Cost and service trade off. For more on that trade-off, see The lowest carrier quote should not always win.

Example: three carriers, one refrigerated lane

Three carriers on one lane show how a gate and a ranking produce a different tender order than a blended score. The figures are illustrative, based on the last 40 loads on a refrigerated lane.

Carrier	Missed reefer pickups (last 40 loads)	Rate vs. target	On-time delivery	Blended score	Two-step result
Carrier X	4	3% below	88%	92	Fails gate (threshold is 2 misses)
Carrier Y	0	2% above	97%	85	Passes, ranks first
Carrier Z	1	At target	91%	87	Passes, ranks second

By blended score, Carrier X gets the next tender. By the two-step method, Carrier Y gets it and Carrier Z is the backup. Carrier X stays approved for dry loads, where its record may be fine.

The method also explains itself. Anyone can read the table and see why Carrier X was skipped. That explanation is what makes a scorecard trusted by the team and by carriers.

How Wilson, by Cartage, uses carrier performance in tendering

Wilson, by Cartage, tenders loads to the company's approved carriers in line with the company's rules, and can produce carrier-performance reports on request. Wilson does not keep a standing carrier scorecard by default. It produces the report when the company asks for one.

The reports come from the loads Wilson runs. Wilson is an AI logistics agent that replaces the TMS for freight execution. It runs the full cycle: quote, book, schedule, track, deliver, audit, and pay. Because it tenders, tracks, and audits each load, its records hold tenders, carrier replies, pickup and delivery events, and invoices.

That gives the company evidence for the measures that matter. Tender acceptance, pickup and delivery timing, and invoice accuracy all sit in the shipment records. Each shipment has a searchable, exportable history with change history.

The company decides what the evidence means. It sets its approved carriers. It sets the rules Wilson tenders by. Wilson tenders to the company's approved carriers first. Cartage can introduce additional carrier capacity, but the company must approve each carrier before Wilson uses it.

When a carrier misses a pickup, Wilson acts on that load. It contacts the carrier and alerts the team. Then it initiates rebooking from the approved carrier list, within the rules or with approval.

To see a carrier-performance report built from your own loads, talk to the Cartage team.

FAQs, conclusion, and sources

FAQs

How should a carrier scorecard influence the next tender?

Use it in two steps. First, a gate removes carriers that fail a requirement for this load. Second, a ranking orders the rest. Wilson, by Cartage, can produce carrier-performance reports on request from the loads it runs, and tenders within the company's rules.

Does Wilson keep a carrier scorecard?

Not by default. Wilson does not maintain a standing scorecard. It can produce carrier-performance reports on request, from the loads it runs.

What is the most important carrier scorecard measure?

It depends on the load. For time-critical freight, on-time pickup is often the gate. For routine freight, rate against target and tender acceptance often rank highest.

How many loads does a carrier need before its score means anything?

Enough to separate a pattern from bad luck. Compare carriers on the same lanes and modes, and treat very short records with caution.

Can a low-scoring carrier still get a tender?

Yes, for a recorded reason. Contract commitments, customer-routed freight, and test loads for new carriers are common examples.

Conclusion

A carrier scorecard earns its place only when it changes the next tender. Use hard gates to protect each load, then rank the carriers that pass. Build every measure from real shipment records. Wilson, by Cartage, tenders within the company's carrier rules and can produce carrier-performance reports on request, so the company can adjust those rules with evidence.