How to Evaluate a 3PL Partner: What to Ask Beyond the Rate Card

Ask what the last serious failure was

A provider who cannot describe a recent failure in specific terms is either unusually fortunate or not being straight with you. Every operation of any size has stopped a customer line, lost stock, or missed a critical dispatch.

The answer is less important than its shape. A useful answer names what happened, how it was detected, how long it took to detect, what the immediate response was, and what changed afterwards. A weak answer is either a denial or a story where the cause was entirely someone else's - a supplier, a carrier, a system.

The detection time is the part worth pressing on. Most operational failures are unavoidable in isolation; what separates providers is whether the problem surfaced within an hour or at the next stock count. That is a function of process discipline rather than of luck, and it is the best available proxy for how your own problems will be handled.

Ask who is actually accountable, by name

The team presenting during selection is frequently not the team running the site. That is not necessarily a problem, but it is worth knowing, because operational quality is largely a function of the site manager and the supervisors, not of the organisation above them.

Ask who will manage the site day to day, whether they are already in post, how long they have been with the business, and what happens when they leave. Ask how many other accounts that person handles. A site manager splitting attention across several customers behaves differently from one dedicated to yours, and neither is wrong, but the difference should be a decision rather than a discovery.

Ask the same about escalation. When something goes wrong at two in the morning, who is called, and what is that person authorised to do without waiting for approval? Escalation paths that require a decision from someone unreachable are escalation paths in name only.

• Who runs the site, and how many other accounts do they carry?
• Who is called out of hours, and what can they authorise unaided?
• What happens to the arrangement if that named person leaves in six months?

Agree what the measures mean before you agree the targets

Most disputes in logistics contracts are definitional rather than performance-related. Both parties report accurately against different definitions and reach opposite conclusions about whether the service is working.

On-time delivery is the classic example. Does the clock start when the order is placed, when it is released to the warehouse, or when it is picked? Is on time measured against the customer's requested date or the date that was confirmed? Are delays caused by the customer excluded, and who decides? Two providers quoting the same target against different answers are not offering the same thing.

The same applies to inventory accuracy, which can be measured by location, by SKU or by value, with or without tolerance - producing very different numbers from the same warehouse. A defined 97 per cent tells you more than an undefined 99, and the definition should be in the contract rather than settled after the first disagreement.

What an accuracy figure has to mean before it is quoted

Ask how the peak is handled, specifically

Every provider will say they can handle your peak. The useful follow-up is how, in terms concrete enough to check.

Where do the additional people come from, and how long have those arrangements been in place? Are they trained on your process before the peak or during it? What is the notice period you must give, and what happens if the peak arrives without notice? Does the provider have other customers peaking in the same weeks - and if so, how is priority decided between you?

That last question is the one providers least like answering and the one that matters most. Shared capacity is the economic logic of outsourcing, and it only works if the peaks are genuinely staggered. A provider whose customers all peak in the same fortnight has sold the same capacity several times.

Look at the transition plan, not just the steady state

Most of the risk in a 3PL arrangement is concentrated in its first eight weeks, and most selection processes barely examine that period.

Ask how stock will be counted and reconciled on day one, and who signs off the opening balance. Ask what happens to material in transit on the cutover date. Ask which system becomes the record and how it exchanges data with yours, and what the fallback is if that interface is not ready. Ask what a rollback looks like if the transition goes badly in week two.

A provider with a considered answer to the rollback question is generally a safer choice than one who has not contemplated it, regardless of how the rest of the comparison looks. The willingness to describe how their own involvement could be unwound is a reasonable proxy for how they will behave when something is genuinely wrong.

How a deployment is sequenced to reduce transition risk

Weigh the rate last, and weigh the structure more

Rate matters, but the structure of the pricing usually matters more, because it determines whose interests move in the same direction as yours.

A price per unit handled rewards a provider for handling more units, which is the opposite of what you want if part of the objective is to reduce handling. A fixed monthly fee removes that incentive and transfers volume risk to the provider, who will price for it. Open-book arrangements with a management fee align the incentives most closely and require a level of trust and administrative effort that not every relationship justifies.

None is universally right. What is universally worth doing is checking which behaviours the pricing rewards, and whether those are the behaviours you actually want in eighteen months.

Where structural cost sits in a logistics network