I get asked to model this perhaps once a month and the spreadsheet is always the same shape, so here it is with a worked example. The company is composite but the numbers are from real engagements.
The scenario
A specialty chemical blender ships 1,800 filled 275-gallon totes a year to forty customer sites, most of them repeat destinations within 600 miles. They currently own a fleet of 520 containers.
Ownership, fully costed
| Line | Cost | Where it currently sits |
|---|---|---|
| Container replacement (11% attrition of 520) | $19,300 | Capital, approved annually |
| Additional units for peak cover (idle 7 months) | $14,100 | Capital — rarely questioned |
| Washing between fills (in-house bay) | $41,400 | Operations labour and utilities |
| Effluent permit and treatment | $6,800 | Facilities |
| Periodic testing and re-certification | $9,600 | Quality, scheduled reactively |
| Empty return logistics | $58,200 | Logistics — largest single line |
| Administration and tracking | $13,900 | Spread across three roles |
| Total | $163,300 |
Annual cost of owning the fleet
Note where those lines live. Capital, operations, facilities, quality, logistics and administration. No single person sees this total, which is exactly why ownership looks cheap.
The pool, same volume
| Line | Cost | Notes |
|---|---|---|
| 1,800 turns at the pooled rate | $124,200 | Includes container, wash, test, float |
| Collection runs, batched by geography | Included | Milk-run rather than per-customer |
| Attrition to the agreed threshold | Included | Above threshold at reclaimed replacement value |
| Above-threshold losses (actual) | $6,400 | Reclaimed value, not new-container cost |
| Asset tagging and reporting | Included | Monthly utilisation and dwell report |
| Internal administration | $8,700 | Reduced but not eliminated |
| Total | $139,300 |
Annual cost of a managed pool
About $24,000 a year, which is a 15% saving and not a dramatic one. The larger effect is elsewhere.
The effect that is not on the spreadsheet
Their fleet was doing 3.5 turns per container per year. The pool runs 6 to 9 because collection is scheduled rather than reactive.
That means the same 1,800 fills need roughly 230 containers in circulation instead of 520. Fewer containers doing more work is the whole point of pooling, and it is where the environmental case lives too — the embodied footprint spreads across three times as many fills.
Ownership is not expensive because containers cost money. It is expensive because owned containers sit still.
Where the dwell goes
| Stage | Owned fleet | Managed pool |
|---|---|---|
| At blender, filled and waiting | 9 days | 6 days |
| In transit to customer | 3 days | 3 days |
| At customer, being emptied | 11 days | 9 days |
| At customer, empty and idle | 54 days | 12 days |
| Return transit | 6 days | 4 days |
| Wash, test, back to float | 21 days | 5 days |
| Total cycle | 104 days | 39 days |
Cycle time breakdown, owned versus pooled
Fifty-four days sitting empty behind somebody's building is the number. It is not anyone's fault — the customer calls when they remember, and the blender collects when a truck happens to be going that way. Scheduled milk-runs fix it structurally.
The environmental arithmetic, since it comes up
Worth doing properly rather than hand-waving, because the result is larger than the cost saving.
| Owned, 3.5 turns/yr | Pooled, 7.8 turns/yr | |
|---|---|---|
| Containers in circulation | 520 | 230 |
| Container-years consumed per 1,800 fills | 514 | 231 |
| Replacement units per year (attrition + wear) | ~57 | ~26 |
| Embodied CO₂e of replacements | ~12.2 t | ~5.6 t |
| Difference | −6.6 t a year |
Embodied footprint per fill, same 1,800 fills
Those figures assume reclaimed replacement units at 150 kg CO₂e each. On new containers the gap would be wider, which is the quiet argument for pooling reclaimed rather than new assets.
When pooling loses
- One-way shipments. If the container is not physically coming back — remote oilfield sites, export, destinations with no outbound freight — pooling is strictly worse. Buy a reconditioned unit and let it go.
- Scattered one-off destinations. A milk-run needs density. Forty repeat sites within 600 miles works; forty different sites a year across the country does not.
- Very low volume. Below about 150 turns a year the fixed administration of a pool outweighs the savings.
- Dedicated containers per product. If cross-contamination risk means each container must stay with one chemistry, a shared pool is the wrong structure.
What to measure before you ask
Two numbers decide this and most companies have neither to hand: turns per container per year, and attrition. Count units purchased against units physically present once, and divide annual fills by containers owned. Those two figures tell you more than any proposal we could write.
The service detail is on tote pooling, and how we count unreturned units is covered in how we count a tote that never comes back.
Written by
Dale Ferreira
Logistics lead
Books every load that leaves the yard. Asks "is there a dock?" in his sleep.