January's Second Peak: Auditing Reverse Logistics as a Designed Service Line
Photo: parcels returns warehouse sorting January logistics, via www.primermagazine.com
The Christmas peak is planned in extraordinary detail: pick rates, temporary labour, carrier capacity, contact centre rosters. Very little of that planning concerns the direction of travel that matters in January. From the first working week, a parallel service line switches on — returns, exchanges, refunds, warranty claims and courier investigations — and in many organisations it is staffed as an exception rather than designed as a service.
The consequences are familiar to anyone who has queued at a parcel counter in early January: refunds that arrive without explanation, contact volumes that spike for reasons the operation cannot name, and a recovery process that leaves value sitting in a warehouse rather than returning to the business.
A service line is defined by its purpose, not its volume
Before auditing anything, write down what the returns service is for. Three purposes compete: recovering value, retaining the customer, and complying with the law. Most operating models have never stated which takes precedence when they conflict — and they do conflict, routinely.
The legal floor is well known in the UK. Distance sales carry a fourteen-day cancellation right running from delivery, with a further fourteen days to return the goods and fourteen for the refund. Many retailers extend this voluntarily over Christmas, and most publish a returns window of twenty-eight or thirty days. Those commitments are service promises, and promises create clocks. An operation that has not designed its clocks against them is relying on the customer not to count.
Intake: returns begin long before the parcel moves
The returns journey starts wherever the customer initiates it: a web form, a contact centre call, a chat conversation, a store counter, or a marketplace dashboard. Each route captures different data, and some capture none.
The audit question is whether the reason for return is coded against a taxonomy that can be used for design. Broad buckets — 'changed mind', 'faulty', 'wrong item' — are better than nothing but rarely survive contact with reality. The useful distinction is between reasons the business can act on and reasons it cannot. Fit and sizing complaints point at product information and photography. Damage points at packaging and carrier handling. 'Not as described' points at the listing itself. Each is a design lever, and each is lost if the reason is typed freehand into a notes field.
Collection and drop-off capacity
The outbound network is modelled weekly. The return network usually is not. In January the shape of it changes: parcel shop queues lengthen, lockers fill, courier collection slots become scarce, and store counters absorb volume they were never designed to handle.
Model the reverse network with the same variables as the outbound one — cut-off times, capacity per day, carrier service levels, local variation. Post Office counters and parcel lockers behave very differently from a home collection, and customers experience the difference as service quality, not as logistics. Where the returns policy says 'free returns' but the only free route is a queue at a counter three miles away, the policy and the experience have parted company.
Refund triggers: the clock the customer actually feels
This is where most of the contact volume originates. Define the trigger event precisely: is the refund clock started at the point of drop-off, at scan into the network, at receipt in the warehouse, at inspection, or at approval? Then measure the elapsed time at each stage.
If goods sit in an inspection queue for five working days, an otherwise compliant refund cycle becomes a borderline one. If the trigger is set at inspection but the tracking token tells the customer the parcel arrived eleven days ago, the customer's mental clock and the operation's clock are running at different speeds, and the gap is filled with contact.
Decide, too, who owns the exception: the item that does not match the order, the return that arrives damaged, the high-value claim requiring review. An exception with no owner becomes a case, and cases accumulate.
Supplier recovery: the value leak
Returned goods have a second journey. They go back to the vendor, into refurbishment, into resale, or into recycling under WEEE obligations for electricals. Each route has a recovery value and a recovery time, and both should be measured.
The audit looks for the reconciliation nobody owns: credits due from suppliers by return, aged beyond thirty days, disputed lines, and returns that have simply left the building without a matching refund or credit. In many operations this is the largest unmanaged number in the whole service, and it never appears in the customer-facing reporting at all. Give it a home in governance with two measures — recovery rate and days to credit.
Communication: three messages and an expectation
Customers do not need a rich journey. They need to know the return was acknowledged, that the goods arrived, and when the money will be back. Where those three messages are missing or generic, contact volume rises in proportion.
Design them as service events with owners rather than as automated niceties. A message that states an expected date and then meets it is worth more than three that express regret.
Turning January into permanent decisions
By the end of February, the evidence is still warm and the appetite for change is at its highest. Take a short pack to the forum that owns service performance: returns volume by reason and channel, cost per return, refund cycle time by stage, recovery rate, contact volume per hundred returns, and the top five exception types by age.
From that pack, three decisions usually follow. Publish the returns rules clearly enough that customers can predict the outcome. Introduce threshold-based collection options where the cost of contact exceeds the cost of collection. And attack returns at source — sizing guidance, product information, packaging — because the cheapest return is the one that never happens.
Seasonal strain is not an argument for temporary fixes. It is the annual stress test of a service line that deserves to be designed once, properly, and reviewed every January with the evidence in hand.