Dispatcher copilot across 200+ daily routes
A regional 3PL ran 200+ daily routes through two dispatchers and a whiteboard. A copilot on McLeod and Samsara cut planning from 2 to 3 hours to under 1 hour.
Call it. An AI answers. That's the demo.
Verify us
- GitHub: Link coming soon
- Clutch: Link coming soon
- Google reviews: Link coming soon
- Crunchbase: Link coming soon
- Press: Link coming soon
Summary
- Client
- Logistics Agency
- Location
- Details to follow
- Modules
- Dispatcher copilot, SMS agent, Email agent
- Integrations
- McLeod, Samsara
- Timeline
- 6 weeks to first deploy
- Commitment and result
- Details to follow
Client anonymized until they approve naming.
A regional logistics agency (LTL and FTL, 40+ shippers) dispatched everything through two senior dispatchers and a whiteboard. We built a dispatcher copilot on McLeod and Samsara drafting the next-day board, then automated check-calls by SMS and inbound shipper email. Route planning per dispatcher fell from 2 to 3 hours to under 1; shipper NPS rose from 32 to 51.
What was the business?
A regional logistics agency operating across the Eastern US, managing LTL and FTL freight for 40+ shippers. $15M to $25M in annual revenue, a fleet of 80+ owner-operators and company drivers, dispatched from a single operations center.
What was broken
All dispatch ran through two senior dispatchers who had been with the company 10+ years. Every driver preference, shipper requirement, and lane rate lived in their heads. When one took vacation, the operation visibly degraded: late pickups, missed appointments, shipper complaints.
Route planning was done by hand on a physical whiteboard and transferred to McLeod each morning. It took 2 to 3 hours per dispatcher per day. In peak season they regularly worked until 8 or 9 p.m. to build the next day's board.
Check-calls were entirely manual. Each dispatcher made 60 to 80 calls a day to drivers and shippers for status updates, time that could have gone to load optimization.
What we built
Dispatcher copilot. An AI dispatch assistant integrated with McLeod and Samsara. It suggests driver-to-load assignments from location, hours available, equipment type, and shipper preferences, and generates the next-day board as a draft for dispatcher review and approval.
SMS agent (check-call automation). Pulls real-time location from Samsara, checks it against pickup and delivery windows, and sends proactive updates to shippers. Exceptions (late, off-route, breakdown) escalate to the dispatchers.
Email agent. Handles inbound shipper email: rate requests, load tenders, POD requests, and status inquiries. Routine requests are handled on their own; complicated ones go to the right person.
Media pending
Media pending
How it went live
- Weeks 1 to 4: deep diagnosis of the dispatch workflow, a McLeod data audit, and driver interview sessions.
- Weeks 5 to 6: dispatcher copilot MVP deployed in suggestion mode only; dispatchers approve every assignment.
- Weeks 7 to 10: check-call SMS automation deployed and the feedback loop established.
- Weeks 11 to 16: email agent deployed; copilot moved to semi-autonomous mode.
- Month 5 onward: ongoing optimization, expanded to include rate quoting assistance.
A custom implementation spanning 6 months: dispatcher copilot, SMS agent, email agent, and custom integration work. It moved to ongoing support at a flat monthly rate.
The numbers
Route planning time per dispatcher
- Before
- 2–3 hours
- After
- under 1 hour (a 50% reduction)
- When
- Mar 2026
- Method
- Route planning time dropped from 2–3 hours to under 1 hour per dispatcher.
Daily routes managed
- Before
- not stated
- After
- 200+
- When
- Mar 2026
- Method
- Details to follow
Manual check-calls per day
- Before
- 60–80
- After
- 10–15 exception-only calls
- When
- Mar 2026
- Method
- Check-calls dropped from 60–80 manual calls/day to 10–15 exception-only calls.
Shipper satisfaction (NPS)
- Before
- 32
- After
- 51
- When
- Mar 2026
- Method
- Shipper satisfaction scores (measured via NPS) increased from 32 to 51 in the first quarter post-deployment.
Dispatcher end of day
- Before
- 8–9 PM
- After
- 5–5:30 PM
- When
- Mar 2026
- Method
- Dispatchers went from working until 8–9 PM to finishing by 5–5:30 PM.
| Metric | Before | After | When | Method |
|---|---|---|---|---|
| Route planning time per dispatcher | 2–3 hours | under 1 hour (a 50% reduction) | Mar 2026 | Route planning time dropped from 2–3 hours to under 1 hour per dispatcher. |
| Daily routes managed | not stated | 200+ | Mar 2026 | Details to follow |
| Manual check-calls per day | 60–80 | 10–15 exception-only calls | Mar 2026 | Check-calls dropped from 60–80 manual calls/day to 10–15 exception-only calls. |
| Shipper satisfaction (NPS) | 32 | 51 | Mar 2026 | Shipper satisfaction scores (measured via NPS) increased from 32 to 51 in the first quarter post-deployment. |
| Dispatcher end of day | 8–9 PM | 5–5:30 PM | Mar 2026 | Dispatchers went from working until 8–9 PM to finishing by 5–5:30 PM. |
In the client's words
To follow with the customer's permission
What would we do differently?
We underestimated how much institutional knowledge lived in the dispatchers' heads about individual driver preferences. The first version of the copilot optimized purely on efficiency metrics and ignored soft factors like "Driver X doesn't do NYC" or "Shipper Y only wants Driver Z." We spent an extra two weeks in month 2 building a preference engine. In logistics engagements since, we capture driver and shipper preferences as structured data during the diagnosis phase.
Verify this
Details to follow
Common questions
What did Plainbuilt build for the logistics agency?
Three modules on McLeod and Samsara. A dispatcher copilot suggests driver-to-load assignments from location, hours available, equipment type, and shipper preferences, and drafts the next-day board for dispatcher review and approval. An SMS agent pulls real-time location from Samsara, checks it against pickup and delivery windows, sends proactive updates to shippers, and escalates exceptions (late, off-route, breakdown) to the dispatchers. An email agent handles inbound shipper email (rate requests, load tenders, POD requests, status inquiries) on its own and sends the complicated ones to the right person.
Which systems does it connect to?
McLeod for the dispatch board and the load data, and Samsara for real-time driver location.
How long did it take to go live?
The dispatcher copilot went live in suggestion mode in weeks 5 to 6, after four weeks of diagnosis, a McLeod data audit, and driver interviews, with dispatchers approving every assignment. Check-call SMS automation followed in weeks 7 to 10 and the email agent in weeks 11 to 16, when the copilot moved to semi-autonomous mode. The implementation spanned 6 months and then moved to ongoing support at a flat monthly rate.