Almost every parcel audit firm sells the same thing: we read your invoices, we find billing errors, we file the claims, we take a cut of what comes back.
That works. Those errors are real. Address corrections that shouldn’t have applied, residential surcharges on commercial addresses, packages manifested and never shipped, duplicate charges, service failures on a guaranteed shipment.
It’s also the smaller number. And after a few years of running that recovery, it gets smaller still.
What refund recovery actually gets you
Two things cap it, and neither is the auditor’s fault.
The claim windows are short. Shorter than most shippers assume, different for each carrier, and different again by claim type — a service-guarantee refund and a damage claim do not run on the same clock, and domestic and international rarely match. Check the current windows in your own agreements and diary them, because they are absolute: miss one and the money is gone regardless of how right you were.
A lot of shippers have already given the refunds away. Service-guarantee refunds are a common bargaining chip in a parcel contract negotiation — you waive the right to claim them, and you get a deeper discount somewhere else in exchange. That’s often a fine trade. But if you signed it and you’re still paying someone a percentage to chase late deliveries, you’re paying for a service that has nothing left to recover.
So the ceiling on recovery is set by your contract and the calendar. Real money, worth collecting, bounded.
The number that isn’t bounded
Here’s the part an invoice audit is not built to find.
A refund is what happens when the carrier charged you something they shouldn’t have. Most of what a parcel account overspends is not that. The carrier charged exactly what was asked of it. The asking was the problem.
Four examples, all of which price correctly and all of which cost you money:
A box that bills more than it weighs. Dimensional weight means you pay for the space, not the contents. Carriers keep tightening how dimensions get measured and rounded, and most shippers have not revisited their carton sizes since the last change. Every shipment in a too-big box is billed correctly and costs more than it should.
A service level nobody chose. Overnight on a package that went across town. Air where ground arrives the same day. Somebody set a default in the shipping system years ago and nobody has looked since. The carrier delivered exactly the service purchased. There is nothing to dispute.
A discount that quietly stopped applying. Parcel discounts are tiered — by weight band, by service, by zone. Your freight profile drifts over three years and the tiers don’t move with it. The effective discount you’re actually realizing separates from the one your contract entitles you to, and nothing on the invoice announces that. Add the minimum charge floor, where the discount simply stops mattering because the package prices at the minimum either way, and a meaningful share of your volume can be earning you no discount at all.
Accessorials that were never negotiated. Residential, delivery area, address correction, fuel. The surcharge stack is where the annual increase lands, and it’s the part of the agreement least likely to have been negotiated in the first place. Auditing it tells you what you paid. It doesn’t change what you’ll pay next month.
None of that produces a refund. All of it produces a bill you could have made smaller.
The one an audit structurally cannot find
Zone skipping is the clearest example, and it’s worth understanding why no parcel-only firm will bring it to you.
If you’re sending two hundred packages a week into the same metro, you’re paying line haul two hundred times. Consolidate them into one LTL shipment into the carrier’s hub in that market, inject them there for final delivery, and you stop doing that.
A parcel auditor can see the pattern in your data. Identifying it is not the hard part. Pricing it is — because pricing it requires an LTL rate, an LTL carrier and somebody who knows what that lane actually costs. A firm that only does parcel has none of those. So the recommendation arrives as “you may want to look into consolidation,” which is not a number, and nothing happens.
This is the one place being able to rate parcel, LTL and truckload on the same platform stops being a convenience and starts being the whole point. A cross-mode recommendation either comes with a real figure attached or it doesn’t get acted on.
Why the good recommendations get ignored too
There’s a failure mode in this category that has nothing to do with analysis quality.
An audit report lands with eight recommendations. The customer acts on three or four. The rest sit there — not because they’re wrong, but because a couple of them were modelled estimates printed next to measured facts, with nothing distinguishing the two. Once a reader hits one number they can’t verify, they discount the whole page. The proven findings go down with the unproven ones.
We build the evaluation the other way round. Every opportunity is labelled by what stands behind it before you see the dollar figure: measured on your own invoices, documented elsewhere but untested here, or modelled from data we don’t hold. Only things measurable on your own bills get bridged into the headline number. Everything modelled sits in a separate register.
The open questions get printed too — what the data can’t answer, what looks wrong in it, and what we chased and abandoned, so nobody rediscovers it next quarter.
That’s not rigour for its own sake. A number that survives your CFO is worth more than a bigger number that doesn’t.
What to actually ask for
If you’re evaluating parcel spend, the questions worth asking are narrow:
- What share of the savings you’re quoting is refund recovery, and what share is something I’d have to change?
- Have I already waived service-guarantee refunds in my contract?
- What’s my effective discount versus what my agreement entitles me to — by service and by weight band?
- What percentage of my volume prices at the minimum charge?
- Which of your recommendations are measured on my invoices, and which are estimates?
- Can you price a cross-mode alternative, or only identify one?
That last one filters the field quickly.
Refund recovery is worth doing. Keep doing it. Just don’t confuse it with knowing what your parcel network costs — and don’t assume the firm collecting a percentage of refunds is motivated to tell you the bigger number sits somewhere else.
If you want the fuller version of the analysis described above, that’s what our parcel practice does — and the same discipline runs on the LTL side as audit while the load is still moving.
So here’s the question worth sitting with: of everything your parcel audit found last year, how much of it required the carrier to be wrong?