Freight cost optimization software: which workflow are you actually buying?
Key Takeaways
Updated November 3, 2026
Freight cost optimization software is a label for four different workflows: freight procurement, rate management, freight audit, and freight execution. Each one saves money in a different place, and most products cover one or two of them. A manufacturer should name the workflow it is buying before it compares tools. An AI logistics coordinator such as Wilson, by Cartage, covers execution, negotiation against rate targets, and invoice audit inside the same load record.
- Procurement buys capacity and sets contract rates. Its output is an award, not a booked load.
- Rate management keeps the right rate on each load: the contract rate when it applies, a negotiated spot rate when it does not.
- Freight audit checks each carrier invoice against what was booked, before payment. Its output is a confirmed or corrected bill.
- Freight execution is the daily work: quotes, booking, pickup confirmation, follow-up, exceptions, documents, and payment. Its output is a load that moved and was paid correctly.
- Money leaks between the workflows: an awarded rate that never reaches the booking, a change that never reaches the invoice check, a carrier that misses pickups and gets rebooked at spot prices.
- Wilson, by Cartage, is an AI logistics coordinator that replaces the TMS for freight execution, negotiates against the company’s rate targets within approval limits, and audits each invoice against the POD and the rate confirmation before paying from the Wilson Wallet.
What are you buying when you buy freight cost optimization software?
Freight cost optimization software can mean a procurement platform, a rate tool, a freight audit service, or an AI logistics coordinator that runs execution. The four workflows have different outputs, different users, and different proofs of savings. The table is the fastest way to tell them apart.
| Workflow | What it decides or does | Output | Who uses it | Where it saves money |
|---|---|---|---|---|
| Freight procurement | Which carriers carry which lanes, at what contract rate | Carrier awards and contract rates | Procurement, transportation leaders | Lower contract rates on steady lanes |
| Rate management | Which rate applies to each load, and whether to negotiate | A rate on each quote or booking | Logistics coordinators, rate desk | Fewer loads paid above contract or target |
| Freight audit | Whether the invoice matches the booked terms and the delivery evidence | A confirmed, corrected, or disputed invoice | Finance, logistics | Overcharges, duplicates, unsupported accessorials caught before payment |
| Freight execution | Quoting, booking, confirming, tracking, exceptions, documents, payment | A load that moved, with its records and a paid invoice | Logistics team, or an AI logistics coordinator | Fewer missed pickups, rebookings, and detention; fewer hours per load |
Hypothetical example: a manufacturer of industrial fasteners ships about 600 LTL and truckload loads a month from one plant. Its CFO asks for “freight cost optimization software.” The logistics manager pulls three months of data and finds four separate leaks: contract rates on two lanes are above market, 11% of loads were booked at spot rates because the contract carrier did not answer in time, detention lines were paid without checking arrival times, and the two coordinators spend most of the day on follow-up. Each leak belongs to a different workflow.
Why the label causes bad purchases
A buyer who hears “cost optimization” often compares a procurement platform with an audit service and an execution tool as if they were the same product. They are not. A procurement platform can win a better rate in a bid and still leave the team booking loads by email. An audit service can recover overcharges and still leave the cause upstream untouched. Naming the workflow first prevents paying for the wrong fix.
The four workflows also hand work to each other. Procurement produces the rate that rate management applies. Rate management produces the booked rate that audit checks. Execution produces the evidence, such as the POD, the arrival times, and the approved changes, that audit needs to judge each line. Software that covers one workflow should show how it receives from, and hands to, the others.
The handoffs between workflows
| Handoff | What has to pass | What goes wrong when it does not |
|---|---|---|
| Procurement → rate management | Awarded carriers, rates, fuel and accessorial terms | Loads get booked at old rates or with carriers outside the award |
| Rate management → execution | The rate and terms on each booked load | The carrier confirms a different price than the one approved |
| Execution → audit | Rate confirmation, approved changes, POD, arrival and departure times | The audit cannot tell a valid charge from an overcharge |
| Audit → procurement | Patterns by carrier and lane, such as repeated detention or declined tenders | The next bid repeats the same problem |
Ask every provider which of these handoffs its software covers, and which it expects another system or a person to handle. The answer tells you more than a feature list.
For a comparison of the AI tools themselves, see the best AI tools for freight cost optimization.
Freight procurement: buying capacity and contract rates
Freight procurement decides which carriers carry which lanes and at what contract rate. An AI logistics coordinator does not replace this decision; the company or its procurement team owns it. The output of procurement is an award: a carrier, a lane, a rate, and terms such as accessorials, fuel, and service expectations.
What procurement software does well
Procurement tools organize a bid: they collect lane volumes, send the request to carriers, compare responses, and record the award. Some add market rate data so the buyer knows whether a bid is high. For lanes with steady weekly volume, a well-run bid can lower the contract rate in a way that holds for months.
Where procurement stops
An award does not move freight. The carrier still has to accept each tender, show up for each pickup, and bill each load at the awarded rate. When the contract carrier declines or does not answer, the load often goes to the spot market at a higher price. Procurement software rarely sees that leakage, because it lives in daily execution.
- An award is not an accepted tender. The carrier can still decline a specific load.
- A contract rate is not a guarantee of capacity. Tender acceptance often drops when the market tightens.
- The award terms, such as fuel and accessorials, need to reach the invoice check, or the audit cannot judge the bill.
Questions to ask a procurement tool
Ask how the awarded rates reach the people or system booking each load. Ask what happens when the awarded carrier declines a tender. Ask how often the tool sees actual tender acceptance. If the answers depend on another system, the buyer needs that other system too.
Hypothetical example, continued: the fastener manufacturer rebids its two high lanes and lowers the contract rate by 6%. In the next month, the contract carrier declines 14 of 60 tenders on those lanes, and the team books those loads at spot rates 18% higher. Part of the procurement saving disappeared in execution. The bid was not wrong; the workflow after it was missing.
When a bid is the right move
A bid pays off when lanes carry steady volume, the current rates are clearly above the market, and the company can commit to the volume it awards. It pays off less when freight is irregular, when most loads move on the spot market, or when the main cost problem is execution: missed pickups, slow tenders, and unchecked invoices. In those cases the bid lowers a number on paper while the actual cost per load stays where it was.
Before a bid, check how often the current contract carriers accept tenders. A carrier that accepts 95% of loads at a slightly higher rate can cost less overall than one that wins the bid and accepts 70%.
Who owns procurement
Procurement usually belongs to a transportation leader, a procurement team, or both. They decide which lanes to bid, which carriers to invite, and how to weigh price against service. Execution data helps them decide better: tender acceptance by carrier, on-time pickup, detention frequency, and the spot premium paid when a contract carrier declines. That data lives in the load records, which is why procurement and execution should share them instead of keeping separate spreadsheets.
Rate management: keeping the right rate on every load
Rate management keeps the right rate on every load: the contract rate when the lane has one, and a negotiated rate against a target when it does not. An AI logistics coordinator can do this load by load, which is where most companies leak money on spot and unplanned freight.
Contract rates first, then negotiation
For lanes with a contract, the job is to tender to the contract carrier first and confirm acceptance quickly. For loads without one, the job is to request quotes from approved carriers, compare them like for like, and negotiate when the best quote is above the target. The target comes from the company: a contract rate, a recent paid rate on the lane, or a budget.
| Situation | Rate rule | What a person decides |
|---|---|---|
| Lane has a contract rate and the carrier accepts | Book at the contract rate | Nothing, if within the rules |
| Contract carrier declines or does not answer in time | Request quotes from other approved carriers | Whether to book above the target |
| Best quote is above target but within the limit | Counter with a specific number, then book | Nothing, if within the approval limit |
| Best quote is above the approval limit | Send the offer and the context for approval | Approve, wait, or change the service |
| New carrier offers the best rate | Do not book until the company approves the carrier | Whether to approve the carrier |
How Wilson manages rates
Wilson, by Cartage, tenders to the company’s approved carriers first and books only carriers the company has approved. When it requests quotes, it negotiates against the company’s rate targets under configured rules and approval limits; rates outside those limits go to the logistics team. Every quote, counter, and approval is saved in the shipment record with change history, so the company can see later why each rate was paid.
Cartage can introduce additional carrier capacity, but the company must approve each carrier before Wilson uses it. Wilson can also produce lane-rate and carrier-performance reports on request from the loads it runs, which helps procurement prepare the next bid with real execution data.
What rate management does not do
Rate management does not design the network or set the contract strategy. It does not guarantee that a carrier will accept a target. Its value is consistency: every load gets the same rules, at the same speed, whether it is the first load of the day or the fiftieth.
Hypothetical example: a contracted lane on a busy week
The fastener manufacturer has a contract rate of $1,850 on a weekly truckload lane. In a busy week, the contract carrier does not answer the tender for four hours. Without follow-up, the coordinator gives up and books a spot carrier at $2,300. With rate management inside execution, the contract carrier is followed up on at the configured interval and called if email does not work. If the carrier declines, Wilson requests quotes from other approved carriers, counters the best one against the $1,950 target, and books at $1,940 within the approval limit. The record shows every step.
The difference is not a better bid. It is the speed and consistency of the work around the tender, applied to every load instead of the loads someone had time for.
Targets need care
A rate target is only useful if it is current. Targets built from last year’s contract or an old spot rate push negotiation in the wrong direction: too low and every load goes to approval, too high and the company leaves money on the table. Review targets by lane and season, and use the rates actually paid on recent loads as one input. When a carrier keeps quoting above target for good reasons, such as a hard-to-serve receiver, change the target rather than fight every load.
The negotiation steps are covered in detail in how AI negotiates freight rates against your targets.
Freight audit: checking the bill against what was booked
Freight audit checks each carrier invoice against what was booked and what actually happened, before payment. An AI logistics coordinator that runs execution has the evidence the audit needs: the rate confirmation, the approved changes, the POD, and the recorded arrival and departure times.
What a useful audit checks
- Linehaul and fuel against the rate confirmation and the contract terms.
- Each accessorial against a written request or an approved change, such as a liftgate, an appointment, or a reconsignment.
- Detention against the recorded arrival and departure times, the appointment, and the free time on the rate confirmation.
- Duplicates, including the same load billed twice under different invoice numbers.
- The POD, to confirm the delivery happened and to note shortages or damage.
Audit after the fact versus audit inside execution
A standalone audit service reviews invoices after they arrive, often with partial evidence. It can recover overcharges, but it usually cannot fix the cause, such as an order that left out a liftgate, because it does not run the booking. Audit inside execution starts with the full record of the load, so a mismatch can be traced to the step that created it.
How Wilson audits and pays
Before paying from the Wilson Wallet, Wilson audits each carrier invoice against the POD and the rate confirmation, and customers can require a signed POD before payment releases. Unsupported detention lines go back to the carrier with the recorded times. If any line does not match what was booked, the whole invoice waits for a person; Wilson does not make partial payments. Payments above approval limits or outside the rules also wait for a person.
Hypothetical example, continued: the fastener manufacturer had been paying detention on every invoice that listed it. With arrival and departure times recorded on each load, three of nine detention lines in a month fall within free time. Those invoices wait for a person, the carrier receives the recorded times, and the corrected invoices are paid. The saving is real, and it is reported as corrected invoices, not as a lower rate.
Accessorials need a written request
An accessorial such as a liftgate, inside delivery, or a residential fee is valid when the company requested it or approved it as a change. A useful audit looks for that request in the load record before accepting the line. When the order left out a known requirement, the charge may be valid even if unexpected; the fix is the order handoff, not a dispute with the carrier.
What audit results mean
| Result | What it means | What to report |
|---|---|---|
| Matched | The invoice agrees with the booked terms and evidence | Paid within the rules |
| Corrected | The carrier sent a revised invoice after a mismatch | The corrected amount, once received |
| Valid but unexpected | The charge is supported, but the company did not plan for it | A process fix upstream, not a saving |
| Open | The question is still with the carrier or a person | Nothing yet; keep it visible |
Only corrected invoices are savings. Flags, open questions, and promised credits are not, until the corrected bill arrives.
For the line-by-line checks, see how AI catches freight invoice errors before payment.
Freight execution: the work that turns a rate into a moved, paid load
Freight execution is the daily work that turns a rate into a moved, paid load: quoting, booking, pickup confirmation, tracking, exception response, documents, and payment. An AI logistics coordinator does this work within the company’s rules, which is where hours, rebookings, and detention costs are won or lost.
Where execution saves money
- Fewer spot bookings, because tenders and follow-up happen fast enough for the contract carrier to accept.
- Fewer missed pickups, because every pickup is confirmed and reconfirmed with the carrier.
- Less detention, because appointments are confirmed and arrival times are recorded.
- Fewer surprise accessorials, because shipment requirements reach the carrier before booking.
- Fewer hours per load, because follow-up, status requests, and document collection no longer depend on a person.
What Wilson does in execution
Wilson, by Cartage, replaces the TMS for freight execution and connects to the company’s ERP. It receives orders through an ERP feed, a scheduled file drop, a spreadsheet, email, or an API. It requests quotes, books approved carriers, confirms pickups, and tracks each load with the best source available: carrier API, GPS or ELD data where it exists, the carrier’s website by PRO number, and email or phone when nothing else answers. It prepares rate confirmations, bills of lading, packing slips, and pallet labels, and collects the signed BOL and POD.
When a pickup is missed, a load is cancelled, or a delivery runs late, Wilson contacts the carrier, notifies the internal team, notifies the customer when configured, and starts rebooking from the approved carrier list, automatically within the rules or after approval. Wilson starts with assisted execution, and routine actions can run automatically within the company’s rules as reliability is proven.
How execution savings should be reported
Execution savings come in two forms: freight dollars (fewer spot loads, less detention, fewer rebookings) and hours (less follow-up). Report them separately. Hours released are capacity, not cash, unless the company changes its spending. Mixing the two produces a savings number nobody can audit.
Execution metrics to track
- Share of contracted loads booked with the contract carrier.
- Time from order release to confirmed booking.
- Pickups confirmed and reconfirmed without a person asking.
- Missed pickups and rebookings per hundred loads.
- Detention lines paid versus detention lines supported by recorded times.
- Hours the team still spends per load, by task.
Measure these on the first lanes against the same period before the change. Use the same definitions before and after, and keep loads that are still open visible instead of dropping them from the report. Most teams start on a few lanes and expand from there; timing depends on scope.
Hypothetical example: a missed pickup on a contracted lane
A carrier misses a 2:00 p.m. pickup at the fastener plant. Without an owner, the dock waits, a coordinator notices at 4:30, and the load ships the next day on a spot carrier at a premium, with a detention charge from the customer’s receiver for the late arrival. With execution covered, Wilson sees no pickup confirmation by the reconfirmation time, contacts carrier dispatch, and calls the driver when the carrier has shared the number. When the carrier confirms it cannot make it, Wilson tells the load owner, requests quotes from approved carriers, and books a replacement within the approval limit the same afternoon. The customer is notified of the new delivery time when configured. The cost avoided is the rebooking premium and the late-delivery charge; both are freight dollars, reported separately from the hours the coordinator did not spend.
How to choose: match the workflow to where the money leaks
A manufacturer should choose freight cost optimization software by finding where its money leaks and buying the workflow that covers that leak. An AI logistics coordinator review of recent loads usually shows the answer within a few weeks of data.
| If the data shows… | The leak is in… | Start with… |
|---|---|---|
| Contract rates above market on steady lanes | Procurement | A rebid with current market data |
| Many loads booked at spot rates on contracted lanes | Rate management and execution | Faster tenders, follow-up, and negotiation against targets |
| Detention, duplicates, or accessorials paid without evidence | Freight audit | Invoice checks against the POD, rate confirmation, and recorded times |
| Missed pickups and rebookings | Execution | Pickup confirmation and exception response from approved carriers |
| Coordinators spending most of the day on follow-up | Execution | An AI logistics coordinator for quotes, follow-up, and documents |
A test that works for any tool
Take ten real loads from the last month, including one that changed after booking, one that missed a pickup, and one with a detention line. Ask each provider to show what its software would do with the same facts. Score four things: the rate obtained, whether the load was confirmed and moved, whether the invoice was checked before payment, and how many hours the team still spent.
- A procurement tool should show how its award reaches the booking.
- An audit tool should show where it gets arrival times and the POD.
- An execution tool should show the full record from quote request to paid invoice.
- Every tool should show what stays with a person, and why.
Warning signs
- A savings estimate before the provider has seen any of your loads.
- One number that mixes lower rates, corrected invoices, and staff hours.
- “Optimization” that means only a dashboard of spend.
- No answer to the question: who books the load at the new rate?
Hypothetical example, concluded: the fastener manufacturer runs the ten-load test. The procurement tool scores well on rate and cannot book. The audit service catches the detention lines and cannot explain the spot bookings. Wilson covers booking, follow-up, negotiation against targets, and invoice audit, while procurement keeps the annual bid. The company buys for the leak it has, not for the label.
Questions to ask in the demo
- Show one of our loads from order to paid invoice. Where does a person act?
- What happens when the contract carrier declines the tender?
- Where does the arrival time used for detention come from?
- How does an approved change reach the invoice check?
- What does the savings report count, and how do you separate rates, corrections, and hours?
- Which of our systems does the software need, and which does it replace?
A provider that answers these with your own loads, rather than a prepared demo, is showing you the workflow you would actually buy.
FAQs
What is freight cost optimization software?
Freight cost optimization software is a label for four workflows: procurement, rate management, freight audit, and freight execution. An AI logistics coordinator such as Wilson, by Cartage, covers execution, negotiation against rate targets, and invoice audit in one load record.
Is freight audit enough to lower freight costs?
No. Freight audit catches overcharges after the load moves, but it does not prevent spot bookings, missed pickups, or detention. An AI logistics coordinator that runs execution fixes those causes and audits each invoice before payment.
Does Wilson replace freight procurement software?
No. Procurement owns contract strategy and annual bids. Wilson, the AI logistics coordinator, applies those rates on every load, negotiates spot loads against targets within approval limits, and can produce lane-rate reports on request for the next bid.
How is Wilson priced?
Wilson has no subscription or platform fee. Pricing is a percentage of the freight Wilson coordinates, with pricing scaling down as volume grows, and Enterprise pricing is custom. An AI logistics coordinator review of recent loads shows which workflow to start with.