CPG Route Optimization: Promotional Volume Buys Vehicles, Not Density

Seasonal peak planning is a solved discipline in CPG. Volume is forecast, capacity is procured months ahead, and the demand curve is matched with a mix of captive, contracted and gig supply. Promotional volume is usually folded into the same exercise as one more input to the forecast.

That works for the capacity number and fails for the route, because a promotional spike and a seasonal spike of the same magnitude have different shapes. Seasonal growth arrives as more orders from more locations. Promotional growth arrives as bigger orders to the same locations. Both raise vehicle requirement in proportion to volume, but only one of them shortens the distance between stops. Plan a promotion with a seasonal model and the vehicle count will be roughly right while the route economics quietly are not.

Why promotional volume is not a small seasonal peak

The commercial exposure is documented. Research by Gruen, Corsten and Bharadwaj across more than 71,000 consumers in 29 countries put the global out-of-stock rate at 8.3%, with the causes breakdown summarized by ECR Retail Loss attributing 38% to store stocking, 22% to store forecasting, 11% each to planning and store ordering and 9% to supply. ECR Europe’s 2003 work found promoted items going out of stock at roughly twice the rate of non-promoted lines. At that ratio, about one in six promoted lines is unavailable during the promotion that was funded to sell it.

The cost never surfaces cleanly. McKinsey surveyed 35 senior leaders at 28 North American consumer packaged goods companies and found only 17% believe they recover more than 75% of the true cost to serve, most estimating about half. Promotional execution cost is a textbook case: it is incurred by transport, funded by trade marketing, and measured by neither.

The unit cost of the answer is rising. ATRI’s operational cost analysis put the industry-average cost to operate a truck at $2.336 per mile in 2025, up 3.4%, with truckload and refrigerated operating margins below 1.0% and carriers leaving 10% of trucks unseated. Incremental vehicles are both more expensive and harder to find than they were.

And the reason the distinction matters at all is that density is the dominant lever in this category. McKinsey’s work on out-of-home delivery finds that raising drops per stop from one to five cuts labor and vehicle cost by more than 50%. Any volume increase that does not improve density is a volume increase that costs full price.

How to plan promotional volume as a routing problem

1. Convert a localised uplift into a network number

A promotion applies to specific accounts, so the network effect depends on both the uplift and the share of accounts covered. Network volume multiple is one minus the covered share, plus the share multiplied by the uplift.

A 3x uplift on a fifth of the account base is a 40% network volume spike. That is peak-season scale arriving on a trade marketing timetable rather than a calendar one.

2. Recognize that the volume lands on stops you already serve

This is the structural difference and it is worth setting out precisely. Take a territory of 500 stops where a vehicle carries 25 units and each stop takes one unit, giving 20 vehicles at 25 stops each.

Both routes to 40% more volume need 28 vehicles. Only the seasonal one gets shorter. Because promotional volume arrives as bigger drops at the same addresses, stop count is unchanged, so the distance between consecutive stops is unchanged, and the density dividend that normally accompanies growth does not appear.

3. Expect stops per vehicle to fall, and plan the consequences

The second-order effect is the one that catches operations out. At 3x on 20% of accounts, vehicles hold 17.9 stops instead of 25. That means more routes covering the same geography, more depot departures inside the same loading window, and more driver shifts to roster, none of which is captured by a volume forecast.

Cost per unit is roughly flat across the two cases, which is why a cost-per-case report will show nothing unusual while the operation runs out of vehicles and dock slots.

4. Use the lead time that already exists

The trade calendar is agreed weeks or months ahead. Sourcing the resulting capacity at short notice throws away a discount you have already earned.

On a 40% spike, covering the increment at same-day rates costs 64% of baseline transport spend against 40% contracted, a waste of 24 percentage points. The multipliers here are illustrative and should be replaced with your own spot-to-contract spread, but the shape holds: the premium is a function of notice, and the notice was available.

5. Treat the calendar as forecastable but not fixed

This is where promotional planning diverges from seasonal planning most sharply. A seasonal peak has a date that does not move. A promotion has a date that is a negotiating position: retailers shift weeks, change mechanics, extend a successful offer or pull a weak one.

So pre-positioning alone captures most of the available benefit and cannot capture all of it. The residual needs the ability to re-plan capacity and routes after the calendar changes, which is a different capability from forecasting well.

6. Model the drop-size change, not just the order count

Promotional orders are usually deeper in fewer SKUs, often full pallets of a single line. That changes how the vehicle fills as well as how much it carries, and a plan built on average drop characteristics will misjudge the load even when it gets the total right.

The effect is sharpest where the promoted line is dense. A mixed baseline load is usually volume-constrained, but a promotion on a heavy category such as beverages, canned goods or liquid detergents concentrates weight, and the vehicle can reach its payload limit with cargo space to spare. The result is a load that looks feasible in a volume-based plan and is not legal on the axles, which is a different failure from simply running short of vehicles. Any promotional plan on a dense category should be checked against payload rather than cube before capacity is committed.

7. Give the promotion an execution owner and a shared metric

The structural fix is organizational. Trade marketing sets the calendar and is measured on sell-through. Transport absorbs the volume and is measured on cost per case. Neither is accountable for whether the promoted stock reached the shelf, which is the outcome both are paid for. A shared measure such as on-shelf availability during the promotional window, reported against the transport plan, is what makes the trade-off visible to both. The practical test of whether that ownership exists is simple: ask who signs off the transport cost of a promotion before the calendar is agreed with the retailer. In most organizations the answer is nobody, which is why the cost is discovered rather than decided.

Where promotional volume differs from other spikes

The middle column is the one most often planned with the left column’s tools. Seasonal peak planning is a mature discipline and is covered in depth elsewhere on this site. The point here is narrower: a promotion is not a small seasonal peak, because the growth has a different shape.

Five criteria for evaluating promotional readiness

1. Does the routing layer receive the promotional calendar as structured data? Not as a forecast adjustment at network level, but as expected uplift by account and by date, which is what the route plan needs.

2. Can it plan at account-and-date granularity rather than region-and-week? A promotion hits specific stores on specific days. Capacity planned at regional weekly level will be right in aggregate and wrong everywhere it matters.

3. Does it model drop size separately from order count? If volume enters the plan only as an order total, the vehicle and dock consequences of bigger drops are invisible until loading.

4. Can capacity be re-planned when the calendar changes? Ask what happens when a retailer moves a promotion by a week after capacity has been contracted, and how much of the plan has to be rebuilt.

5. Is on-shelf availability during the promotional window reported against the transport plan? Without that link the execution cost and the commercial outcome stay in separate reports owned by separate teams.

What this looks like in enterprise deployments

A global FMCG manufacturer distributing across ten Asian countries through more than 1,000 distributors and 5,000 riders reached 3X ROI while saving more than 12,000 trips a month and optimizing over $4 billion of orders across 1.8 million retail outlets. Trips saved is the right measure for this problem specifically. When volume arrives as bigger drops rather than more stops, the vehicle and the trip are the scarce units, and consolidation that removes a trip is worth more than routing that removes a mile.

A leading North American retailer running multi-hundred stores across ocean, rail and road replaced six legacy systems with a single orchestration layer, reaching over 99% on-time store delivery, route compliance above 95% and exceptions resolved in under two hours. On-time store delivery is the metric that carries a promotion: promoted stock that arrives after the offer starts has already lost most of its value, and 99% reliability is the difference between a funded promotion and a funded disappointment.

Four mistakes in planning promotional volume

Folding the promotion into the seasonal forecast. It gets the tonnage roughly right and the route structure wrong, because the growth has a different shape.

Sourcing promotional capacity at short notice. The calendar gave weeks of warning. Paying spot rates for volume you knew about is the most avoidable cost in the category.

Planning at regional weekly granularity. Promotions hit named accounts on named days. Aggregate capacity that is adequate for the week can be short on the two days that matter.

Measuring transport on cost per case during a promotion. Cost per case barely moves when volume arrives as bigger drops, so the metric reports normality while vehicles and dock slots run out.

How Locus handles forecastable, account-specific volume spikes

Locus, the world’s first Decision-Intelligent, Agentic TMS, is built for the case where the plan has to change after it was made, which is what a negotiable calendar creates. The route planning system sequences against more than 250 real-world operating constraints, holding vehicle capacity by weight and volume together with time windows, store receiving constraints and driver hours, so a promotional drop-size change is evaluated as a capacity condition rather than discovered at the dock.

Capacity Management predicts and manages fleet and driver capacity, plans around shifts and absence, and engages third-party carriers for surges, which is the machinery for converting a known calendar into contracted rather than spot capacity. Allocation then runs across owned fleet, contracted transporters and a network of more than 1,000 carriers, decided per shipment against live cost and serviceability, so the incremental capacity a promotion needs can be sourced by route rather than by blanket commitment.

Because plans are produced in roughly two minutes and re-optimized continuously, a promotion that moves by a week does not require the plan to be rebuilt by hand. That responsiveness is the part pre-positioning cannot supply on its own, and it is the reason forecasting accuracy and re-planning capability are complements rather than substitutes.

Across FMCG and CPG deployments the recurring pattern is roughly 15% lower freight cost, 20% fewer vehicles required, 84% less planning time and 95% SLA attainment, with route planning delivering up to 34% fewer miles, 25% higher drop density and 28% fewer trips through order consolidation. On promotional volume the vehicle reduction and the trip reduction are the figures that matter, because those are the units the spike consumes.

Locus supports more than 360 enterprise customers across 30-plus countries, with over 1.5 billion deliveries optimized, more than $320 million in documented client logistics savings and 99.99% uptime. It has been recognized by Gartner for seven consecutive years, featured in the 2026 Hype Cycle for Supply Chain Execution and Logistics Technologies, named a Leader in TMS by QKS Group (SPARK Matrix), and ranked #1 in Route Planning on G2’s 2026 Best Software Awards.

In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Locus continues to operate independently.

So how should CPG route optimization handle promotional volume? As a spike with a different shape from a seasonal one, not as a smaller version of it. A 3x uplift on a fifth of accounts is a 40% network volume event, and because it arrives as bigger drops at stops you already serve, it raises vehicle requirement in proportion to volume while leaving distance per stop unchanged, where the same 40% arriving as new stops would have shortened distance per stop by about 15.5%. Stops per vehicle falls from 25 to about 18, so the binding constraints become vehicles, dock slots and driver shifts rather than tonnage, and cost per case reports normality throughout. The lead time to fix this is already in the trade calendar, and sourcing that capacity at same-day rates wastes around 24 percentage points of baseline transport cost. Locus models drop size as a capacity condition inside a 250-plus constraint set, converts a known calendar into contracted capacity through Capacity Management and a 1,000-plus carrier network, and re-optimizes in roughly two minutes when a retailer moves the date. Request a Locus route planning assessment to model your own promotional volume shape.

Frequently Asked Questions

Is a promotional spike the same as a seasonal peak? No, and treating it as one is the common error. Both raise volume and both raise vehicle requirement in proportion. The difference is shape: seasonal growth arrives as more orders from more locations, which shortens the distance between stops, while promotional growth arrives as bigger orders to existing locations, which does not. A 40% rise delivers about 15.5% shorter distance per stop in the seasonal case and none in the promotional case.

How does a promotion on a few accounts become a network problem? Multiply the uplift by the share of accounts covered. A 3x uplift on 20% of accounts produces a 40% network volume increase, and a 5x uplift on 30% produces 120%. The promotion is negotiated account by account and lands on the network as a single capacity event.

What actually runs short during a promotion? Vehicles, dock slots and driver shifts rather than tonnage. Because volume arrives as bigger drops, stops per vehicle falls, in the worked example from 25 to about 18, so the same territory needs more routes departing inside the same loading window. Cost per case stays roughly flat, which is why the constraint is often invisible in reporting.

Why is promotional capacity so often bought at spot rates? Because the promotional calendar lives in trade marketing and reaches transport late, if at all. The notice exists, it simply is not transmitted. On a 40% spike, covering the increment at same-day rates costs about 64% of baseline transport spend against 40% contracted, a waste of roughly 24 percentage points for information that was already agreed weeks earlier.

If the calendar is known, is pre-positioning enough? Not on its own, because a promotional date is a negotiating position rather than a fixed event. If 20% of promotions move or change mechanic after the calendar is agreed, pre-positioning captures 80% of the available benefit and the remainder needs the ability to re-plan capacity and routes after the change. Forecasting and re-planning are complements here.

Do promotions increase out-of-stocks? Yes. ECR Europe’s 2003 work found promoted items going out of stock at roughly twice the rate of non-promoted lines, against a global base rate of 8.3% from the Gruen, Corsten and Bharadwaj study. At that ratio roughly one in six promoted lines is unavailable during the promotional window, which is the commercial cost of getting the execution wrong.