A routing guide should execute an approved capacity policy, including eligibility, tender deadlines, fallback order, and controlled exceptions.

A spreadsheet that lists three carriers and a rate is not a routing guide. It becomes useless at the first decline, timeout, pickup change, or shipment that does not fit the lane's standard assumptions.

A working routing guide is an executable capacity policy. It tells the shipping operation:

  • whether a shipment is eligible for the guide;
  • which service and equipment may be used;
  • which approved carrier receives the first tender;
  • how long that carrier has to respond;
  • what happens after a decline or timeout;
  • who can buy outside the contracted waterfall;
  • how much they may spend;
  • what must be revalidated before freight is released; and
  • which events and costs are recorded for review.

The most important freight routing guide best practice is to design the failure path before launch. “Call transportation if primary declines” is not a fallback policy; it is a queue with no clock.

The decision rule near the top

For each shipment, execute this sequence:

  1. Validate the shipment. Confirm the origin, destination, ready time, appointment, weight, dimensions, commodity, handling attributes, equipment, value, and payer.
  2. Apply hard eligibility gates. Exclude providers that lack the approved authority, insurance, geography, equipment, commodity, customer, or facility fit.
  3. Select the service. Apply the mode and service rule before selecting a carrier.
  4. Rank eligible capacity. Use the current lane and service hierarchy—not a global carrier rank.
  5. Tender one controlled offer. Start the response clock and preserve the tendered terms.
  6. Advance on decline or timeout. Cancel or close the prior offer, record the reason, and tender the next eligible provider.
  7. Escalate when the contracted path is exhausted. Use a defined spot, expedite, or service-change approval.
  8. Reconfirm identity and terms at pickup. An accepted tender does not authorize an unverified performing-carrier or destination change.

That sequence can run in a TMS, an API workflow, EDI, or a disciplined manual process. Automation is useful only when the policy and data are explicit.

Build the routing guide on a complete shipment record

A carrier cannot accept what the tender does not disclose. Required fields should include:

Field group Minimum useful data Failure caught
Parties and locations Legal shipper, origin, destination, bill-to, exact facility and contacts Wrong site, wrong payer, ambiguous instructions
Timing Ready time, pickup window, delivery appointment, transit commitment, time zone Physically impossible or misunderstood service
Freight Commodity, pieces, pallets, weight, dimensions, stackability, value Wrong equipment, capacity, handling, or coverage
Service Mode, equipment, team/solo, drop/live, temperature, securement, stop count Tender sent to a carrier that cannot perform
Regulated attributes Hazmat details, food or pharmaceutical controls, permits, high-value procedures Ineligible carrier or illegal/unsafe execution
Commercial terms Contract/rate reference, fuel version, included services, authorized accessorials Invoice dispute or underpriced exception
Control identifiers Shipment ID, PO/order, tender version, seal or logistics-unit identifiers Duplicate, stale, or unauditable instruction

Do not route on fields that the source system does not reliably maintain. If dimensions or freight class are frequently missing, the control is not “required” merely because the TMS displays a red asterisk. Define who corrects the data, by when, and what happens if it remains unknown.

Standard logistics-unit identity can help connect physical freight to the electronic instruction. GS1 describes the SSCC as a unique logistics-unit identifier that can link scans to electronic business messages in its Logistic Label Guideline. Use it where trading partners support it; do not make a standard label a substitute for correct commodity and handling data.

Separate eligibility from carrier preference

The primary carrier should be the first eligible provider, not simply the name in row one.

Eligibility rules can include:

  • active and appropriate operating authority where required;
  • insurance verified for the commodity and exposure;
  • approved carrier agreement and rate;
  • equipment, permits, endorsements, and cargo capability;
  • service geography and facility access;
  • customer-specific approval;
  • temperature, food, hazmat, high-value, or securement controls;
  • enough feasible driver time and lead time for the appointment;
  • no active suspension, safety stop, fraud hold, or payment hold; and
  • available capacity for the tender date.

Connect these gates directly to the motor-carrier vetting process. A carrier's approval should be scoped by mode, equipment, commodity, geography, and risk. The routing engine should not infer that “approved for dry van” means approved for every tender under the same USDOT number.

Preference comes after eligibility. Rank providers by the service and commercial policy for that lane: contracted commitment, expected acceptance, service performance, total cost, claims or exception performance, facility fit, and strategic allocation. Do not allow a lower rate to offset a failed hard gate.

Design the tender as a state machine

Every tender should have one state and an event history:

State Required event Next permitted state
Ready Shipment validation passed Tendered or exception hold
Tendered Offer sent with version and response deadline Accepted, declined, timed out, or canceled
Accepted Authorized carrier accepted the stated shipment Dispatched, canceled, or exception review
Declined Carrier declined with reason Next eligible tender
Timed out No valid response by deadline Cancel/close and tender next
Dispatched Driver/equipment assigned and identity confirmed At pickup or recovery
At pickup Facility verified carrier, equipment, and release conditions In transit or rejected
Recovery Accepted capacity failed or shipment changed materially Retender, service change, or cancellation

This prevents silent double tendering. If the primary times out and later accepts after the secondary has been awarded, the system must know which acceptance is valid and send an unambiguous cancellation to the other carrier.

Electronic messages do not solve state ambiguity by themselves. X12 defines the 204 Motor Carrier Load Tender for offering a full-load motor-carrier shipment with schedule, equipment, commodity, and instruction data. X12 states that it is not the LTL bill-of-lading, pickup-notification, or manifest transaction. The 990 Response to a Load Tender is the response to a 204. Review the X12 transportation transaction-set descriptions before assuming one implementation fits every mode.

X12 also distinguishes routing information from shipment authorization: the 853 Routing and Carrier Instruction can communicate detailed routing and carrier information but does not itself convey a shipping event or authorize shipment. Preserve that distinction in custom APIs and portals as well.

Whether the response arrives by 990, API, portal, email, or telephone, record:

  • tender version;
  • sent and received timestamps;
  • accepted, declined, or timed-out state;
  • person or system that responded;
  • reason code;
  • price and service accepted;
  • any qualifications or changed terms; and
  • the carrier and authority under which service will be performed.

An acceptance that changes the pickup date, equipment, rate, or performing entity is a counteroffer or exception, not a clean acceptance.

After dispatch, the 214 Transportation Carrier Shipment Status Message can communicate dates, times, locations, routes, identifiers, and conveyance details. Whether status arrives by 214 or another channel, define the required events, event clock, source timestamp, and late-or-missing-event response. A stream of location pings does not replace a controlled exception state.

Set response clocks from the operating problem

There is no defensible universal tender timeout. A scheduled contract lane tendered three days ahead can use a different clock from a same-day recovery load. Build response times from:

  • lead time to pickup;
  • likelihood and cost of rejection;
  • depth of approved backup capacity;
  • time needed for carrier dispatch;
  • facility appointment and cancellation rules;
  • time zone and operating hours; and
  • whether the shipment can tolerate a service or mode change.

Use response-time bands, for example “planned,” “next-day,” and “same-day recovery,” then tune them from actual acceptance data. The exact minutes are a shipper policy, not an industry fact.

Make clocks visible to both parties. The tender should state the expiration. The next carrier should not lose feasible lead time because an unanswered primary tender remained open indefinitely.

Build a real fallback ladder

A useful waterfall contains more than Primary–Secondary–Tertiary:

  1. Committed primary capacity for the lane and service.
  2. Approved contracted backup with the same service eligibility.
  3. Approved alternate mode or service, only if the shipment's cost and promise allow it.
  4. Approved managed or spot-capacity channel with the same identity, authority, insurance, and pickup gates.
  5. Exception desk with a rate ceiling and named approval authority.
  6. Business decision to change the appointment, split or consolidate the shipment, expedite, hold, or cancel.

For every rung, define:

  • entry condition;
  • eligible providers;
  • response clock;
  • pricing authority;
  • who can change service;
  • customer-notification requirement;
  • security and identity checks;
  • required reason code; and
  • terminal stop condition.

Do not let “spot” mean unvetted. Urgency increases identity-fraud exposure because teams are more willing to accept changed contacts, unfamiliar domains, or carrier substitutions. Apply the controls in the freight-fraud control system at the point of award and again at pickup.

Test whether the plan is physically executable

Driver time and dwell

For property-carrying commercial motor vehicles, federal hours-of-service rules generally include an 11-hour driving limit within a 14-hour driving window after 10 consecutive hours off duty, subject to detailed provisions and exceptions. FMCSA's Interstate Truck Driver's Guide to Hours of Service is a useful operational reference.

Do not convert a mileage calculation into a service promise without pickup dwell, appointment time, parking, traffic, team/solo assumptions, and the driver's available hours. The carrier owns driver compliance; the shipper owns issuing feasible instructions and not designing a guide that routinely depends on unsafe or noncompliant recovery.

Size, weight, and route

Federal truck size and weight rules do not create one unrestricted national route. FHWA explains that federal weight standards apply on the Interstate System, while states may set standards off it, and that size standards apply on the designated National Network. Review the FHWA commercial vehicle size and weight overview.

Oversize, overweight, hazmat, tunnel, bridge, local-delivery, and facility-access constraints belong in eligibility and routing—not in free-text notes discovered after dispatch. Placarded hazmat moves add requirements under 49 CFR Part 397; FMCSA's hazardous-materials compliance overview points operators to the controlling regulations.

Weather, closures, and operating restrictions

A static route is not evidence that the route is currently open. Establish the information source and check owner for severe weather, closures, chain controls, construction, and emergency restrictions. FHWA points travelers to state and national road-condition systems through its real-time traveler information resources. The carrier remains responsible for safe vehicle operation; the shipper's guide should define how service and customer commitments change when the route does.

Preserve the commercial logic

Each routing-guide rate needs:

  • carrier and legal entity;
  • lane, direction, and valid geography;
  • mode, equipment, service, and shipment assumptions;
  • base rate and minimum;
  • fuel method and source;
  • included stops, dwell, or handling;
  • accessorial rules and approval;
  • effective and expiration dates;
  • volume or commitment assumptions;
  • tender-lead and cancellation terms; and
  • contract and tariff hierarchy.

Version the guide. A shipment should retain the rate and policy version used when it was tendered. Replacing a spreadsheet or table in place destroys evidence for invoice audit and performance review.

The related freight-pricing architecture and detention, demurrage, per-diem, and storage clocks should feed these terms. A routing guide that chooses a linehaul rate but ignores likely accessorials is not choosing total cost.

Define exceptions before they happen

Create reason codes and owners for:

  • no eligible carrier;
  • all carriers declined or timed out;
  • shipment data incomplete;
  • load changed after acceptance;
  • carrier fell off after acceptance;
  • driver or equipment mismatch at pickup;
  • rate exceeds authority;
  • appointment no longer feasible;
  • route or weather restriction;
  • cargo, packaging, or securement issue;
  • customer-directed carrier or service; and
  • system outage.

Each code should trigger a response, not merely improve reporting. A performing-carrier mismatch should hold release. A rate over the desk's authority should require approval. A material shipment change should invalidate the old tender and launch a controlled retender.

Provide a manual continuity procedure for TMS, EDI, or API outages. It should preserve the same approved providers, decision order, response record, and authority limits. “Use email until the system returns” is incomplete unless the team knows which email, template, version, approval, and reconciliation step apply.

Measure the routing guide without rewarding bad behavior

Useful measures include:

  • first-tender acceptance by lane, lead-time band, and carrier;
  • acceptance and decline reason by tender position;
  • median response time and timeout rate;
  • carrier falloff after acceptance;
  • routing-guide compliance;
  • percentage of shipments reaching spot or exception desks;
  • rate and total-cost variance to the eligible first choice;
  • on-time pickup and delivery;
  • detention and avoidable accessorials; and
  • identity, authority, or shipment-data holds.

Define routing-guide compliance carefully. If the primary was ineligible or had already declined, using the secondary is compliant. Comparing the invoice carrier only with the first name in the static guide will misclassify correct execution.

Review tender performance with carriers. A low acceptance rate may reflect poor carrier execution, but it may also expose unrealistic volume commitments, late tenders, dirty shipment data, wrong lane definitions, or rates that do not cover the operating requirement.

Launch by breaking the guide on purpose

Before broad rollout, test:

  • primary declines;
  • primary never responds;
  • late primary acceptance arrives after backup award;
  • shipment weight or appointment changes after acceptance;
  • backup carrier has an insurance or authority hold;
  • performing truck does not match the tendered carrier;
  • rate exceeds the exception ceiling;
  • road closure makes the appointment infeasible;
  • EDI or TMS is unavailable; and
  • customer changes destination in transit.

Confirm the event, owner, clock, communication, approval, and audit record for each. Pilot on a bounded lane set, reconcile every tender state and invoice, then expand.

The operating takeaway

A freight routing guide should make the routine move automatic and the abnormal move controlled. Its quality is visible when the first carrier says no: the next eligible action is immediate, authorized, and recorded without weakening carrier, cargo, or payment controls.

Anchor the guide in the broader operator's map of how freight moves across America, then make each mode-specific tender and handoff explicit.

Use the UniteCargo carrier directory to develop candidate depth, then contract, vet, and scope each provider before it enters the waterfall. For programs managed through an intermediary, first resolve whether the provider is acting as broker, 3PL, or freight forwarder.

Sources and further reading