Wilson by Cartage
Try Wilson →
Freight pickups

A carrier scorecard should change who gets the next load

Key Takeaways

  • A carrier scorecard is only useful if it changes who gets the next load. A quarterly report that nobody checks at booking time doesn't change carrier behavior or results.
  • 5 measures matter at the moment of assignment, by lane: on-time pickup, on-time delivery, response time, invoice accuracy and rate versus the lane benchmark.
  • Scores have to be current and come from your own shipments. A score built from last quarter's spreadsheet can't stop this afternoon's booking.
  • Wilson, by Cartage, is an AI logistics agent that applies the company's carrier and approval rules when it quotes and books, and it can produce carrier-performance reports on request from the loads it runs, to help the team update those rules.
  • People still own the carrier list and the weighting. The company decides which carriers are approved and how much service counts against price.

Why do carrier scorecards often fail to change who gets the next load?

Carrier scorecards often fail to change load assignment because they're built after the fact, in a separate spreadsheet, by someone other than the person booking. At booking time, the coordinator sees quotes in an inbox, not scores. Unless the score sits next to the quote when the decision is made, the fastest reply or the lowest rate wins.

This affects manufacturers and distributors that manage their own carriers and review them quarterly, often from incomplete records. Visibility gaps make it worse: in a March 2025 survey of 240 supply chain leaders, BCI Global found only 35% reported end-to-end visibility. When pickup and delivery times are pieced together from emails, the scorecard is incomplete before anyone reads it.

One lane, one week

For example, a distributor has three approved carriers on a Chicago–Nashville LTL lane.

  • Monday 9:00 a.m. The quarterly review shows Carrier C missed 6 of 20 pickups on the lane (for example). The logistics manager notes it in the slide deck.
  • Tuesday 2:10 p.m. A new order lands in the ERP. The coordinator sends a quote request to the usual three carriers. A quote is the carrier's price for that shipment.
  • Tuesday 2:45 p.m. Carrier C replies first, $40 under the others. The coordinator, who wasn't in Monday's meeting, books it and sends the rate confirmation (the document recording the agreed rate, accessorials and pickup window).
  • Wednesday 4:30 p.m. The pickup window closes with no truck: a missed pickup. The coordinator calls the dispatcher, who promises Thursday.
  • Thursday 10:00 a.m. Customer service fields the customer's complaint, and the coordinator rebooks with another carrier at a short-notice rate.

The scorecard was right. It just wasn't anywhere near the decision.

What should a carrier scorecard measure before the next load?

Before the next load, a scorecard should show 5 measures for that lane, from your own recent shipments. Each one predicts a different kind of cost. Response time is the one most teams skip; see how to measure carrier response time.

Measure What it shows Where the data comes from
On-time pickup Whether the truck arrives in the window Pickup confirmations and check calls
On-time delivery Whether the load arrives as promised Tracking and the POD (proof of delivery, the signed delivery receipt)
Response time How fast the carrier answers quotes and status checks Email and call history
Invoice accuracy How often the invoice matches the booking Invoice audit against the rate confirmation
Rate vs. benchmark How the rate compares with the lane Booked rates on the lane

Two conditions make these measures fair. First, score by lane, not network-wide: a carrier can be reliable out of Chicago and weak out of Atlanta. Second, require a minimum number of loads before a score drives decisions, for example 10 loads on the lane, so one bad week doesn’t bench a good carrier. For how price and service trade off, see your cheapest carrier may be your most expensive decision.

How should scores change load assignment?

Scores should change load assignment through a few written rules that apply at booking time, not through a quarterly conversation. The rules should be simple enough that a coordinator or an AI agent can follow them on every load.

Use the score as a tie-breaker when rates are close. For example, if quotes are within 3%, the carrier with the better on-time pickup on that lane gets the load.

Cap volume below a set score. A carrier under the threshold on a lane gets fewer loads there until it recovers.

Require approval for low-scoring carriers on key freight. Key customer loads go to a low scorer only with a manager's sign-off.

Review carriers whose scores drop. A conversation with the carrier often fixes a dispatcher or terminal issue before volume has to move.

Add approved carriers on lanes where every score is low. That's a capacity issue, not a carrier issue.

Each rule has a consequence. Caps move volume to other carriers, which may raise the average rate on the lane. Approval rules add a step on key loads. Teams should decide those trade-offs once, in writing, rather than load by load.

Common mistakes

  • Building the scorecard quarterly. It describes the past and never meets the next booking.
  • Scoring network-wide. A carrier strong overall keeps winning a lane where it fails.
  • Scoring from too few loads. One bad week benches a good carrier, and the lane loses capacity.
  • Never telling the carrier. The carrier can't fix what it doesn't know, so the score doesn't improve.
  • Treating the score as a ranking only. Without a booking rule, the score changes nothing.

How does Wilson, by Cartage, use carrier performance data when it books?

The company defines the carrier and approval rules, such as preferred carriers per lane, tie-break rules, volume caps and approval on key loads, and Wilson, by Cartage, applies those rules when it quotes and books. Wilson doesn’t maintain a standing carrier scorecard by default, but it can produce carrier-performance reports on request from the loads it runs, to help the team update the rules. Traditional logistics software gives teams information. Wilson executes the work required to move shipments forward.

Step Trigger What Wilson does When a person steps in
1. Record Each shipment runs Records pickup confirmations, tracking updates, the POD and the invoice audit on the order Not required
2. Quote The next shipment is created Requests quotes from the approved carriers, starting with the company’s preferred carriers for the lane Not required
3. Negotiate Quotes arrive Negotiates against the company’s targets Not required
4. Book The best option is clear Books under the company’s rules, such as tie-break rules and volume caps, and sends the rate confirmation Key loads, carrier selections or rates set to need approval
5. Report The team asks for a report Produces a carrier-performance report from the loads it has run The team reviews it and updates the rules

Because Wilson runs the tender, the tracking, the POD collection and the invoice audit on the same load, a carrier-performance report it produces on request is built from the work itself rather than from a spreadsheet assembled later. The team can use it to update the booking rules, and Wilson applies the updated rules from the next load on.

The company sets the controls. It decides which carriers are approved on each lane, how much service weighs against price and which carrier selections need approval. Wilson starts with assisted execution, and every message, confirmation and change is recorded in a change history, so the team can see why a load went to one carrier and not another.

Wilson books only carriers the company has approved; it doesn’t add or remove carriers on its own. For more on how an AI logistics agent works across the shipment, see what is an AI logistics agent.

What should stay with a person?

People should own the carrier relationship, the approved list and the weighting between service and price. Those are judgment calls with commercial consequences.

The logistics manager decides which carriers are approved per lane and when a carrier is removed. The team decides how much on-time pickup is worth in dollars on key customer freight. Conversations with carriers whose scores drop belong to the person who owns that relationship. When a lane needs more capacity, Cartage can introduce carriers from its network, and they’re available to Wilson only after the company approves them. See best carrier scorecard software for distributors for how categories of tools compare.

What to ask a vendor to show in a demo

  • Show one of your carriers’ scores on one lane, and the shipments behind it.
  • Show what the coordinator, or the system, sees at booking time: quotes and scores side by side.
  • Set a rule that a low-scoring carrier on a key customer load needs approval, then trigger it.
  • Show where a missed pickup and a mismatched invoice are recorded against the carrier.
  • Show the change history explaining why a load went to one carrier over a cheaper one.

FAQs

  • How should a carrier scorecard change load assignment? Through written rules applied at booking time: use the score as a tie-breaker when rates are close, cap volume for low scorers and require approval on key loads. With Wilson, by Cartage, the company defines those rules and Wilson applies them when it quotes and books.
  • What should a carrier scorecard measure? On-time pickup, on-time delivery, response time, invoice accuracy and rate versus the lane benchmark, by lane, from your own recent shipments.
  • How often should carrier scores be updated? As each load completes. A quarterly score describes the past and can't influence the next booking.
  • Does Wilson provide carrier scorecards? Not as a standing scorecard by default. The company defines the carrier and approval rules, such as preferred carriers per lane, tie-break rules, volume caps and approval on key loads, and Wilson applies those rules when it quotes and books. On request, Wilson can use shipment and rate data from the loads it runs to produce carrier-performance reports that help the team update the rules.
  • Does Wilson always book the best-performing carrier? Not necessarily. Wilson negotiates against the company's targets and books the best option under the company's rules and approval limits, such as tie-break rules and volume caps, which the company sets.
  • Who decides which carriers are approved? The company. Wilson only books carriers the company has approved, and Cartage can introduce carriers from its network for the company's approval.
  • How is Wilson priced? There's no subscription fee. The company preloads a wallet, Wilson pays carriers from it once each invoice is audited and reconciled, and pricing is a percentage of those payments, covering the full shipment lifecycle. Pricing scales down as volume grows, and Enterprise pricing is custom.

Conclusion

A scorecard that sits in a quarterly deck doesn't change anything. The one that matters is current, scored by lane from your own shipments, and tied to written rules that apply when the next load is assigned. That's when a carrier that keeps missing pickups stops getting the freight.

Wilson, by Cartage, applies the company's carrier and approval rules when it quotes and books, with every decision in the change history, and can produce carrier-performance reports on request from the loads it runs to help the team update those rules. Bring the lane where a low-scoring carrier kept getting loads last quarter and watch Wilson quote and book it under your rules in a demo at cartage.ai.

Sources