Operations & Supply ChainLogistics & Warehousing

AI Prompts for Logistics Cost Analysis

Logistics cost is easy to total and hard to explain. A freight bill tells you what you paid; it does not tell you which customers, order profiles, lanes or decisions caused it. Cost-to-serve analysis allocates logistics cost to the things that drive it — order size, delivery frequency, distance, mode, accessorials — so you can see which parts of the business are expensive to serve and why.

These prompts build that allocation, analyze the drivers, and produce a ranked list of savings opportunities. They work from your invoice, shipment and order data; the model should show the allocation logic and its weaknesses rather than present allocated costs as precise.

Before you use these

Have these ready to replace the highlighted [variables]:

Freight and handling costs by shipment, lane, mode and carrier for the period
Shipment data: weight, volume, distance, origin/destination, service level, accessorial charges
Order data: order size, lines, frequency by customer or channel
Warehouse handling costs and activity counts

1. Build the cost-to-serve breakdown

Best forAllocating logistics cost to customers, channels and order types so the expensive ones become visible.

Inputs needed

Cost data
Shipment and order data
Allocation preferences

How to use itGive the model the activity data it needs to allocate (shipments, lines, pallets) not just the totals. Ask it to state where allocation is a proxy.

Expected outputCost-to-serve table by customer/channel with cost per order, per unit and as % of revenue, plus the allocation logic and its limits.

Act as a logistics analyst building a cost-to-serve model for [business / period].

Costs: [freight by shipment or lane; warehouse handling (receiving, picking, packing) with activity counts; returns handling; accessorials]
Activity data: [by customer or channel: orders, lines, units, shipments, pallets/parcels, weight, distance, delivery frequency, returns]
Revenue by customer/channel: [if available]

1. Define cost pools and drivers: e.g. freight → weight-distance or shipment; picking → lines; packing → parcels; receiving → pallets; returns → return units. State the driver for each pool and why.
2. Compute driver rates (cost per driver unit) and allocate to customers/channels.
3. Present cost-to-serve per customer/channel: total, per order, per unit, per revenue $. Rank by cost as % of revenue.
4. Identify the profile features that make a customer expensive: small orders, high frequency, remote location, expedited service, high returns, special handling.
5. Compare the most and least expensive customers with similar revenue and explain the difference through the drivers.
6. State the weaknesses of the allocation (e.g. shared truck cost split by weight ignores volume; fixed warehouse cost allocated by activity overstates the marginal cost of small customers) and the customers whose figures are least reliable.

Do not present allocated cost as marginal cost. Show the arithmetic for one customer end to end.

2. Analyze the cost drivers

Best forUnderstanding why cost moved and what would change it, rather than just where it sits.

Inputs needed

Cost trend
Shipment profile changes
Rate changes

How to use itGive period-over-period data. The model should separate rate effects from volume and mix effects.

Expected outputDriver decomposition of cost change (rate, volume, mix, service level, accessorials), the largest controllable drivers, and the questions for carriers and internal teams.

You are analyzing logistics cost drivers for [business] between [period 1] and [period 2].

Data: [by period: total cost by mode/lane; shipments; weight; volume; average shipment size; mode mix; service-level mix (standard/expedited); accessorial charges by type; carrier rates or contract changes; fuel surcharge]

1. Decompose the cost change into: rate effect (same volume at new rates), volume effect, shipment-size effect (cost per unit changes with consolidation), mode/service mix effect, accessorial effect. Show the method and the figures.
2. For each effect, state whether it is external (rates, fuel), structural (network, customer mix) or operational (consolidation, expedite decisions, order cut-offs).
3. Identify the largest controllable drivers and the process or decision behind each (e.g. expedites caused by late production; small shipments caused by order cut-off times; accessorials caused by address data).
4. Quantify the cost of the top three controllable behaviors per period.
5. Questions to ask carriers (rate basis, minimums, accessorial triggers) and internal teams (why expedites occur, order consolidation rules).

Present as a bridge from period 1 to period 2 cost with each effect labeled. Flag where the data does not allow a clean separation.

3. Rank the savings opportunities

Best forA prioritized list of logistics savings with value, effort and risk stated.

Inputs needed

Cost-to-serve and driver findings
Constraints (service commitments, contracts)

How to use itGive the model the constraints. Savings that break a customer's service level are not savings.

Expected outputOpportunity list with savings estimate and basis, implementation effort, service and risk effects, owner and sequence.

Act as a logistics manager building a savings program from a cost-to-serve and driver analysis.

Findings: [expensive-to-serve segments, controllable drivers, rate observations]
Constraints: [customer service commitments, carrier contract terms and end dates, warehouse capacity, systems]

Generate opportunities across: consolidation (order cut-offs, delivery frequency, minimum order rules); mode shift; carrier procurement and rate benchmarking; accessorial elimination (address quality, appointment scheduling, packaging); expedite reduction (root causes upstream); network changes (stock positioning, cross-dock); customer-facing changes (minimum order value, delivery day scheduling, charging for expedite); packaging and cube utilization; returns process.

For each: annual savings estimate with the basis and confidence, one-off cost and effort, effect on service and on which customers, risk, owner, time to realize, and dependencies.

Then:
1. Rank by savings net of risk, and separate quick wins from projects.
2. Sequence: what to do in the next quarter and what needs contract or system changes first.
3. Identify opportunities that require a commercial conversation with customers, and draft the framing for that conversation.
4. Total addressable savings with a range.

Do not count the same shipment's savings under two opportunities. State the assumption behind every estimate.

Related prompts

Logical next step

After this, most operations teams move on to Transportation Planning.

Get the free Operations & Supply Chain AI Starter Kit → Nine of these prompts as a diagnose → analyze → plan workflow with an intake worksheet, delivered by email. See what's inside

All Operations & Supply Chain prompts · Search the full library