A single 3PL had grown with the plant. Rates had been extended, not tested. Finance could see the invoice total. Operations could not see a clean cost per consignment. Peak surcharges and accessorials were treated as climate. They were habits.
The difficulty
Outbound freight sat with one operator. There was no baseline that separated line-haul from exception charges. Dispatch data and invoices did not join. When management asked whether the 3PL was competitive, the only answer available was the 3PL's own dashboard. That is not an answer. It is a brochure.
What we changed
We rebuilt a twelve-month baseline from invoices and dispatch, separated exception codes, and wrote a demand pack the market could price. A shortlist was scored on cost, service geography, systems and transition risk. The incumbent was invited. They were not protected. The client ran the commercial negotiation. We stayed out of the contract.
Difficulty
One uncontested relationship. No cost per consignment. Data sat with the supplier.
Measure
Invoice baseline, demand pack, scorecard, contestable shortlist.
Result
A defendable baseline, a recommended set, an SLA draft the client owned.
The result
Management received a number they could take to a board without asking the 3PL to explain it. Exception cost became visible enough to manage. Whether they kept the incumbent or moved is their commercial decision and is not published here. The method is the point: you cannot contest what you cannot see, and you cannot see what you have never extracted from your own invoices.
The handover was a workbook, a one-page decision and an SLA draft. Operations could run the new measurement on a Monday. That is what a cost file is for.
What the invoices actually showed
Once line-haul was separated from exception, the story changed. A share of the spend was waiting time at the plant, not carrier greed. A share was redelivery after a customer window the 3PL had never been given in a usable form. Peak had been treated as climate for three seasons. Cubic rules were punishing a packaging habit the plant had never been asked to change. None of that would have been visible in a rate-only tender. A cheaper base rate would have left the leakage in place and called it a win in month one.
The demand pack therefore included exception history, not only tonnes. Shortlisted operators were asked to price the work as it actually ran, including the messy days. The incumbent's knowledge of those days was an advantage they were allowed to use. It was not a reason to skip the contest. Scoring weighted cost, geography, systems access and the right to a volume extract the manufacturer would hold.
What operations could do after handover
Cost per consignment became a weekly number, not a quarterly argument. Exception codes had owners. The SLA draft named clocks, not adjectives. Whether the plant kept the incumbent remains confidential. Management could take a baseline to a board without asking the 3PL to explain the 3PL.