Rate cards hide the costs that matter most in 3PL fulfillment. These nine performance metrics — with target ranges and contract language — tell you whether a provider can really carry your volume.
Choosing a third-party logistics provider on price per order is a bit like choosing a surgeon on hourly rate. The number is knowable, comparable, and almost entirely beside the point. What determines whether a 3PL relationship works is operational performance — and performance is only manageable if you agreed on how to measure it before you signed.
Why Rate Cards Mislead
Two providers can quote nearly identical pick-and-pack rates and produce wildly different total costs. The variance hides in dimensional weight rules, storage billing methods, receiving fees, special-project charges, and — most expensively — the cost of errors you end up absorbing.
A reshipped order does not just cost the second shipment. It costs the original unit, the support time, the refund exposure, and some fraction of a customer relationship. At a 2% error rate on 10,000 monthly orders, that is 200 incidents a month quietly eroding a rate advantage that looked like a win in the spreadsheet.
The Nine Metrics Worth Writing Into a Contract
Accuracy and quality
- Order accuracy rate — orders shipped with correct items, quantities, and shipping addresses. Target 99.5%+; below 99% is a structural problem, not bad luck.
- Inventory record accuracy — matched locations at cycle count. Target 99%+.
- Damage rate — units damaged in storage or outbound handling, as a share of units shipped.
Speed
- Same-day ship rate — orders received before cutoff that ship the same day. Target 97%+.
- Dock-to-stock time — hours from container arrival to inventory being sellable. Anything beyond 48 hours will distort your buying and your listings.
- Average click-to-door — total elapsed time from customer order to delivery, which is what the customer actually experiences.
Cost and reliability
- Cost per order, fully loaded — including storage, receiving, and surcharges, divided by orders shipped.
- On-time delivery rate — measured against the carrier promise, segmented by service level.
- Return processing time — days from carrier receipt to restocked or dispositioned.
| Metric | Healthy | Watch | Escalate |
|---|
| Order accuracy | ≥ 99.5% | 99.0–99.5% | < 99.0% |
| Same-day ship | ≥ 97% | 93–97% | < 93% |
| Dock-to-stock | ≤ 24 hrs | 24–48 hrs | > 48 hrs |
| Inventory accuracy | ≥ 99% | 97–99% | < 97% |
| Return processing | ≤ 3 days | 3–7 days | > 7 days |
Make the Metrics Enforceable
A metric without a definition, a reporting cadence, and a consequence is a wish. In the service level agreement, pin down:
- Exact definitions. Does "same-day ship" mean a label was printed, or that the carrier scanned it? These produce different numbers.
- Measurement source. Whose system is authoritative, and can you audit it against your own order data?
- Cadence. Weekly dashboards, monthly business reviews, quarterly commercial reviews.
- Remedies. Credits for sustained misses, a defined cure period, and a termination right if the cure fails.
- Exclusions. Carrier delays, weather, and your own late inventory should not count against the provider — but the exclusions must be listed, not improvised.
Questions That Reveal More Than a Tour
- What is your order accuracy for the last twelve months, by month, not as an average?
- What percentage of your clients are within 20% of my volume? Being an outlier in either direction is risky.
- How does peak season change staffing, and what did last November's accuracy look like?
- What does your WMS expose through API, and how quickly does inventory sync to my storefront?
- Who is my day-to-day contact, and what is their escalation path at 4pm on a Friday?
Costs That Appear After Onboarding
Budget for these explicitly rather than discovering them:
- Receiving billed per carton, pallet, or hour — the unit matters enormously for small-carton freight
- Storage by pallet, bin, or cubic foot, with long-term surcharges on slow movers
- Kitting and special projects at hourly rates
- Account minimums that penalize seasonal volume troughs
- Offboarding — removal fees and notice periods that can make leaving expensive
Run a Structured Pilot
Before full migration, move a representative slice of SKUs — including your awkward ones — and run for 60 days. Measure the nine metrics yourself from your own order data rather than accepting the provider's report. Send test orders. Send a test return. Send an order with a problem and watch how support handles it.
A pilot that costs you two months of dual operations is dramatically cheaper than a failed migration during peak. The providers worth partnering with will welcome the scrutiny — they know their numbers, and their numbers are how they win.