Profitable contract logistics

A logistics provider sells a service whose volume it does not control. The customer decides how many orders it sends, at what time, in what form and with what requirements. The provider, meanwhile, has committed to a price.

This tension lies at the heart of the business and explains most of its difficulties: contracts that are profitable on paper and doubtful in practice, services performed but never invoiced, and customers whom providers neither dare leave nor manage to make pay.

This document offers a method for regaining control. It assumes no change in price: the first step is simply to know what you actually do.

What is your margin on customer X?

Almost every logistics provider’s management team answers this question with an average. Total revenue, total costs, a ratio. That is management-accounting information, not operational management information.

What you would need to know, and what is almost always missing:

• how many hours were actually spent on this customer last month;
• how many square metres it actually occupied, day by day;
• how many out-of-contract services were performed for it;
• which of them were invoiced.

Without these four data points, margin by customer is a hypothesis. And a company that does not know its margin by customer unknowingly makes its good contracts subsidise its bad ones.

Ask your operations team. The number of immediate answers shows where you stand.

• Which customer requested the most exceptional services from us this month?
• How many pallets belonging to customer Y did we store on the 12th of the month?
• How long does it take us to pick an average order for customer Z?
• Which services did we perform last week without an associated invoice line?
• What is our cost of processing a receipt, across all customers?

Zero or one answer: invoicing is reconstructed from memory, and leakage is significant.

Two or three: you have data, but it is not consolidated.

Four or five: your difficulty is not measurement; it is negotiation.

What you actually invoice

A logistics service breaks down into a small number of activity units. The list below covers most contracts.

The penultimate category deserves more detail, because it is the main source of lost margin.

Special labelling, co-packing, kitting, enhanced quality control, repacking, returns, point-of-sale display preparation, short-dated stock management and allocation by store.

These services share three characteristics:

• they are requested verbally, often urgently and often by an operational contact at the customer;
• they are carried out immediately, because refusing would damage the relationship;
• they are not linked to any document when they occur.

Three weeks later, at invoicing time, someone has to remember. The big ones are remembered. The small ones — two hours here, half a day there — fade away.

The reasoning that loses the most money in this business fits into one sentence: ‘it is not much, we are not going to invoice that’. Repeated fifteen times a month on one contract, it represents a significant share of the expected margin.

The rule: declare the service when it is performed

The solution is not to invoice more. It is to know what you did.

The rule: every out-of-contract service is declared when it is performed, by the person performing it, with three pieces of information — which customer, what type of service, and what duration or quantity.

What happens next is a commercial decision made by someone who has the mandate to make it:

• invoice at the agreed rate;
• invoice with a concession;
• do not invoice, but know it and be able to show it to the customer at the annual review.

The third option often deserves long-term consideration. A table showing a customer what was delivered outside the agreed package over twelve months provides a useful factual basis at renegotiation time.

The objection is always the same, and it is legitimate: teams do not have time to enter data.

Three principles make declaration sustainable:

A closed list. Service types are predefined for each customer and drawn from the contract. No free-text field: choose from a list of eight to fifteen lines.

Three actions at most. Customer — type — quantity. Everything else (date, time, operator, site) is inferred automatically.

Where the work happens. On a terminal in the operating area, not in an office at the end of the day.

If declaration takes more than twenty seconds, it will not be done on busy days — precisely the days when there is most to declare.

PART THREE — THE PRICE SCHEDULE FOR EACH CUSTOMER

Every customer has its own rules: activity units, volume-discount thresholds, minimum charges, fixed fees and negotiated exceptions. A provider serving fifteen customers manages fifteen price schedules.

A usable price schedule includes, at minimum:

CUSTOMER · [name] · CONTRACT from [date] to [date]

INVOICING FREQUENCY · [monthly / fortnightly / other]

STORAGE

Unit · [pallet-day / m²-month /...]

Rate ·...

Minimum charge ·...

Volume discount · from... → rate...

Calculation basis · [average stock / peak stock / stock on a fixed date]

INBOUND FLOWS

Unit ·... Rate ·... Included in the fixed fee? · yes/no

OUTBOUND FLOWS

Unit ·... Rate ·... Minimum order threshold ·...

VALUE ADDED (closed list of services that can be declared)

[type 1] · unit · rate

[type 2] · unit · rate

FIXED FEES

SERVICE COMMITMENTS

[indicator] · threshold · consequence

SPECIAL CLAUSES

The storage ‘calculation basis’ line causes the most disagreements. Invoicing average stock, peak stock or stock on a fixed date produces three very different amounts for the same service. If the contract does not specify it, you and the customer will interpret it differently.

Many contracts include commitments — service level, availability time, stock accuracy — with penalties.

Two rules of caution.

Never commit to an indicator that you do not measure yourself. Otherwise the customer will measure it using its own definition, and you will discover the penalty at the same time as the calculation.

Write down the exclusions. An order submitted after the cut-off time, a product unavailable because of the customer, an incorrect address: these cases must be excluded from the calculation, and that must be written down. Otherwise you will be penalised for failures that are not yours.

A warehouse serving several principals must ensure separation at four levels. Each has its own cost and risk.

Stock. Goods belonging to two customers never mix, even when the item is identical. This sounds obvious and is regularly breached in practice, particularly during re-slotting.

Data. Customer A sees only its own data. This applies to its online access, reports and exports.

Rules. Each customer has its own rotation method, checks, labelling and picking method. The warehouse applies different rules according to who owns the goods.

Evidence. Each customer must be able to obtain traceability for its own flows without seeing anyone else’s.

The third level is the one most often neglected. A warehouse applying one operating procedure to all customers makes real operational savings — and gives up the ability to contract for specific requirements, which means the ability to charge more.

The customer portal: a commercial issue disguised as a technical one

Giving a customer online access to its stock and orders is presented as a service. Above all, it reduces workload: every self-service consultation is one email your operations team does not have to handle.

But it is also a commitment. Once the customer can see its data live, it can also see your variances, delays and errors, without filters or explanations.

A portal should therefore expose only data for which you can answer. Opening access to stock whose accuracy is not under control turns a service into a source of disputes.

PART FIVE — ONBOARDING A CUSTOMER

A prospect choosing between two providers compares prices. But the decision is often won by the answer to another question: how soon can you start?

A provider able to open a new customer in weeks rather than months can shorten mobilisation time and respond more effectively to urgent transitions. The economic value of that advantage depends on the contract and implementation costs.

This lead time depends very little on the tool. It depends on what was prepared in advance.

Before signature

• ☐ Activity units defined and priced
• ☐ Storage calculation basis written down
• ☐ Closed list of value-added services
• ☐ Service commitments with their exclusions
• ☐ Forecast volumes: items, movements, seasonality
• ☐ Product constraints: temperature, batches, dates, hazardous properties

Before the first flow

• ☐ Order-exchange format defined and tested
• ☐ Item master received and checked
• ☐ Locations assigned
• ☐ Rotation and control rules configured
• ☐ Document templates approved by the customer
• ☐ A lead contact appointed on both sides
• ☐ Emergency procedure: who calls whom outside business hours

First week

• ☐ Daily check-in with the customer
• ☐ Variances recorded and handled within 24 hours
• ☐ First out-of-contract services declared from day one

The last point is the most important and the most neglected. The habits of the first week become the habits of the contract. If exceptional services are not declared at the start, they never will be.

PART SIX — WHAT CHANGED OUR MINDS

We once advised starting with invoicing: that is where the money is visible and where management expects a result.

We now start by capturing services, even when invoicing remains manual for several months.

The reason comes from a repeated observation: automating invoicing from incomplete data produces incorrect invoices faster. The provider loses the customer’s trust with the first automated invoices, and the project takes a year to recover.

The sequence that works: capture, verify for two or three cycles, then automate. The first automated invoice must be identical to the one that would have been produced by hand — that is the only test that matters.

SELF-ASSESSMENT GRID

Do we have a closed ☐ ☐ list of services for each customer?

Do we measure ☐ ☐ the indicators to which we commit ourselves?

Can we apply ☐ ☐ different operating rules by customer?

Do we know how long ☐ ☐ it takes to onboard a new customer?