Dispatcher Accountability: The Weekly Load-Review Scorecard That Closes the Leak

Dispatchers don't lose money because they're lazy. They lose it because nobody is grading their work in a structured way, week over week, with the numbers on the table. Software shows you what was logged. A scorecard shows you what was decided.

Across the carriers we've embedded with, the single highest-ROI operational change — bigger than a new TMS, bigger than a rate negotiation cycle — is the introduction of a weekly load-review scorecard. It takes 45 minutes. It closes 60–70% of dispatcher leakage within two quarters.

Why dispatchers drift

A dispatcher's day is 10 hours of micro-decisions: which load, which driver, which lane, which accessorial to fight for, which to let go. None of those decisions are visible individually. Collectively, they are your margin. Without a feedback loop, drift is the default — every shortcut becomes the new normal within a quarter.

The five KPIs that actually matter

Skip the 30-metric dashboard. Five numbers, per dispatcher, per week. If a number can't be tied directly to a decision the dispatcher made, it doesn't belong on the scorecard.

• Revenue per truck per day (RPTD). The cleanest single measure of how hard a dispatcher's board is working. Target by lane mix; benchmark against the dispatcher's own 4-week trailing average.
• Deadhead percentage. Empty miles ÷ total miles, on loads the dispatcher booked. Healthy is under 12%. Over 15% is a planning failure, not a market failure.
• Accessorial capture rate. Of the detention, layover, TONU, and lumper events that occurred, what percentage actually got billed? Under 70% means money is on the floor.
• On-time service score. Pickup + delivery on time, weighted equally. Anything under 96% costs you the next contract renewal.
• Driver home-time hit rate. Requests honored within 24 hours of the date promised. Drives retention more than pay does after the first 90 days.

The scorecard format

One sheet per dispatcher. Five columns: KPI, this week, 4-week trailing, target, variance. Two columns of context: top three loads (by revenue) and bottom three loads (by margin). Nothing else. The whole document fits on a single page and takes ten minutes to fill out from your TMS.

What makes it work isn't the format — it's that the dispatcher fills it out themselves before the meeting. Self-scoring forces them to look at the same loads you're about to look at. By the time you sit down, half the conversation has already happened in their head.

The 45-minute ritual

Same time every week. Friday morning works best — the week is closed, the weekend hasn't started, and decisions about next Monday haven't hardened yet. The structure:

1. Numbers (10 min). Walk the five KPIs. No discussion yet. Just the numbers, this week vs. trailing, vs. target.
2. Top three loads (10 min). What worked, why it worked, what's repeatable. Catching wins matters as much as catching misses — it's how a dispatcher learns the pattern.
3. Bottom three loads (15 min). The hard conversation. Each load gets the same four questions: what was the call, what was the alternative, what would you do differently, what changes for next week.
4. One commitment (10 min). The dispatcher names one specific behavior change for the next seven days. Written down. Referenced in the next meeting. No vague "I'll do better on accessorials" — it has to be "I will submit detention requests within 4 hours of driver notification, every time, and Sarah will spot-check three per week."

What kills it

Three things sink the scorecard ritual every time, and they are the same three things in every shop we've seen:

• Skipping weeks. Once you skip one, you skip the next. The compounding signal is that this isn't actually important. Run it 50 weeks a year, no exceptions.
• Turning it into a punishment meeting. If the bottom-three review feels like an interrogation, dispatchers stop being honest about their losing loads. They'll quietly bury them. You need calm, structured, repeatable — not punitive.
• Owner not in the room. The first 12 weeks, the owner runs it personally. After that, an ops manager can take it — but only after the standard is set. Delegating it on week one signals it doesn't matter.

What to expect

Weeks 1–4 are noisy. Dispatchers are uncomfortable, the numbers are dirty, and the conversations are awkward. By week 6 the rhythm settles. By week 10 you'll see the first measurable lift in accessorial capture — usually 8–15 percentage points. By the end of the second quarter, deadhead drops, RPTD climbs, and the leakage curve bends.

The scorecard is not the magic. The ritual is. The dispatchers know that every Friday morning, somebody who can read a P&L is going to sit across from them and walk three of their worst loads. That is what changes Monday's decisions.