Building a cost-per-trip model
Almost every expensive pallet decision is made by comparing unit prices. Eight lines of arithmetic produces the number that actually matters, and it frequently reverses the answer.
Unit price is what people argue about; cost per trip is what they pay
Ask us to model yours
Give us annual volumes, return rate if you know it, and what you currently pay. We will send the arithmetic — including the case for changing nothing, which is roughly one audit in five.
- Inputs
- 8
- Deciding figure
- Return rate
- Typical saving found
- 12–24%
- Time to build
- An afternoon
A cheaper pallet that lasts one trip is more expensive than a dearer pallet that lasts six. Everybody knows this and almost nobody calculates it.
The eight inputs
- Delivered unit cost. Not the quoted price — add the freight to your dock, because pallets are mostly air and freight is a large share of the real number.
- Useful trips per unit. Ask your supplier, then halve it until you have your own data. This is the input people are most optimistic about.
- Return rate. Of the units leaving your building, what share comes back into your control? Operations assume 80% and discover 45%.
- Repair rate and repair cost. What proportion needs work per cycle, and what each intervention costs.
- Handling labour. Count the touches each unit takes inside your building, multiply by minutes, multiply by fully loaded labour cost. Nobody books this to pallets and it is rarely small.
- Disposal, or credit. If you pay a hauler to remove broken units this is a cost. If you sell them to a yard that pays, it is revenue — and that sign change alone reverses some comparisons.
- Failure cost. Rare, expensive, and the whole reason cheapest-per-unit is not cheapest-per-trip. Include an expected value, however rough.
- Then divide. Total cost across the asset's life, divided by useful trips.
Return rate is the whole model
If you only measure one thing, measure this. It determines whether pooling, fleet ownership or buying recycled is right, and it is the input that most often inverts a conclusion. Above roughly 85 percent return, a durable asset amortises properly and both pooling and ownership become viable. Between 50 and 85 percent you are in the messy middle, where pool loss charges bite and owned fleets bleed invisibly. Below 50 percent your pallet is packaging and should be bought and priced like it.
Measuring it needs one field on the bill of lading recording units out, plus a monthly zone reconciliation. Half a day a month, no software, and the first three months will look wrong before the real number appears.
The worked comparison
The table in our technical reference runs four options side by side using mid-2020s Central Texas figures. The headline: a repaired recycled unit at six trips comes out around $5.72 per trip, a new build at ten trips around $6.47, a one-way recycled unit around $14.60, and pooled rental anywhere between $8.50 and $12.40 depending entirely on return rate.
Note what that says. The new build is not much worse per trip than the repaired unit — its longer life nearly pays for its higher price. What makes it worse is capital outlay and loss exposure, both of which are return-rate questions again.
Four places the saving usually hides
Specification above the job
Paying for Grade A cosmetics on units destined for a muddy site or a one-way journey. The commonest expensive habit in this trade, and the easiest to fix.
One-way freight both directions
A trailer delivers graded units and leaves empty; another trailer later collects empties. Two movements for one visit's worth of work. Two-way accounts price materially better because the trailer was going to be there anyway.
Paying to dispose of inventory
Broken units are the raw material for a repair bench. If you are paying a hauler to take them, you have found the easiest cost saving in your building.
Handling touches
Each one damages a small percentage. An operation with four internal touches destroys roughly four times as many units as one with a single touch, and the staging layout usually causes more of them than the equipment does.
What to do with the number
Negotiate on it rather than on unit price. A supplier who can move your cost per trip — by grading to a written standard so you stop over-specifying, by collecting on the same movement that delivers, by paying for your broken stock, by reducing your handling — is worth more than one who shaves forty cents off a unit. And a supplier who cannot discuss the number at all is telling you something.
Specification appendix
The four things worth putting in writing for any sector, and the reference tables behind them.
How to specify, in ten lines
- Footprint, length first, in inches — and the carton pattern that produced it, so we can check for overhang.
- Construction: stringer or block, two-way or four-way entry. Specify block if you cannot control racking orientation.
- Maximum deck gap in inches. This is the number that decides whether a carton falls through your rollers, and no grade implies it.
- Minimum entry height if you run powered pallet jacks. Repairs with thicker boards quietly eat clearance.
- Squareness tolerance. ±¼" for racking and automation; ±½" is fine for floor storage and cheaper.
- Moisture ceiling, with the measurement method named. 19% standard, 15% for food contact or chilled service.
- Treatment: HT and ISPM-15 marked, and re-marked after any repair introducing untreated wood.
- Species, if stiffness matters. Mixed hardwood for racked service; southern yellow pine where nothing spans anything.
- Permitted repairs: plates allowed or not, companion boards allowed or not. Automation usually wants neither.
- Load case: weight, support condition, dwell time, whether it stacks. This is what lets us pick the grade rather than you guessing it.
Grades
| Grade A | Grade B | #1 premium | #2 economy | Custom cut | |
|---|---|---|---|---|---|
| Original boards | All | Some replaced | All | Several replaced | Mixed |
| Companion boards | None | Permitted | None | Permitted | Permitted |
| Stringer plugs | None | Permitted | None | Permitted | Permitted |
| Metal plates | None | Permitted, disclosed | None | Permitted | Permitted |
| Squareness | Within ¼" | Within ½" | Within ¼" | Within ½" | Within ½" |
| Colour uniformity | High | Mixed | High | Mixed | Mixed, fresh saw ends |
| Protruding fasteners | Zero | Zero | Zero | Zero | Zero |
| Moisture ceiling | 19% (15% on request) | 19% | 19% (15% on request) | 19% | 19% |
| Typical service | Retail-facing, export | Internal, bulk | Automation, racking | One-way, yard storage | Odd footprints |
| Relative price | Highest recycled | −18 to −30% | High | Lowest | ≈⅓ of a new build |
| Remaining life | Longest | Shorter — already repaired once | Longest | Shortest | Varies |
Grade A and #1 overlap: we use #1 for units that needed no repair at all and A for units that are cosmetically uniform. Where a customer holds their own written standard, that document governs and supersedes these names.
Tolerances
- Overall footprint
- ±¼" is normal in recycled stock. ±⅛" is achievable on new builds and worth paying for only if you run automation.
- Squareness
- Diagonal difference under ½". Beyond that, racking and ASRS reject it, and stretch wrap tightens on two corners and slackens on the other two.
- Deck gap
- Typically 3½" between boards. Conveyor lines often need a maximum of 3". Specify the number in inches — it is not implied by a grade.
- Entry height
- Nominal 3½". A repair with thicker boards can lose ¼" and still look fine, then wedge a powered pallet jack.
- Board thickness
- Nominal ⅝" may measure 0.60–0.66". Specify a minimum if deck flatness matters to your product.
- Protruding fasteners
- Zero tolerance. Any nail standing above the deck surface condemns the unit at our bench.
- Moisture content
- 19% is the mould threshold and our standard ceiling. 15% on request for food contact and chilled service.
- Bark
- Some export destinations require debarked wood. Ask before you load, not after.
Load ratings
| Construction | Static | Dynamic | Racked (unsupported) | Unit weight |
|---|---|---|---|---|
| Recycled GMA 48×40 stringer, sound | 4,600 lb | 2,800 lb | 1,500 lb | 33–45 lb |
| Recycled 48×40 block, sound | 5,200 lb | 3,000 lb | 1,900 lb | 40–55 lb |
| Recycled 48×40 stringer, repaired (B) | 4,400 lb | 2,650 lb | 1,400 lb | 35–48 lb |
| New southern yellow pine, 5 top boards | 4,200 lb | 2,500 lb | 1,400 lb | 33–38 lb |
| New mixed hardwood block, 7 top boards | 6,000 lb | 3,500 lb | 2,400 lb | 48–58 lb |
| Heavy-duty new build, doubled stringers | 9,000 lb | 5,000 lb | 3,800 lb | 62–78 lb |
| Skid (no bottom deck) | 3,800 lb | 2,200 lb | Not for racking | 24–32 lb |
Representative planning figures for sound stock, not engineering certifications. Racked ratings assume the load spans two rails with nothing under the middle. Derate a further 20–30% for dwell times beyond a few weeks, because wood creeps under sustained load.
- Design ceiling: deflection of span ÷ 100 — about 0.4 inches across a 40-inch unsupported span.
- Orientation: rack a stringer pallet across the wrong axis and the deck boards carry the span instead of the stringers. Capacity can halve.
- Creep: wood deflects progressively under sustained load. A unit measured inside tolerance on day one can exceed it after six months in a rack.
- Stiffness, not strength, is the limit. Mixed hardwood outperforms southern yellow pine in racking by more than their strength figures suggest, because it is substantially stiffer.
- Drive-in racking gives a long unsupported span with no support at the leading edge. Derate hard, use block construction, consider full-perimeter bottom boards.
Moisture
| Moisture content | What happens | Typical source | Freight weight |
|---|---|---|---|
| 6–12% | Indoor equilibrium. No mould risk. Dimensionally stable. | Kiln dried, stored inside | Lightest |
| 12–19% | Safe working range — our standard ceiling. | Covered dry stack | Normal |
| 19–25% | Mould within days to weeks in warm conditions. | Rain-exposed stacks, green lumber | +5–10% |
| 25–40% | Mould rapid, blue stain likely. | Pallets left in the open through a wet week | +10–20% |
| Above 40% | Saturated. You are paying freight on water. | Standing water, flooded yard | +20%+ |
Mould needs free water. Below roughly 19% the water in wood is chemically bound in the cell walls and spores cannot germinate; above it, they can. Heat treatment kills insects and confers no moisture guarantee; kiln drying reduces moisture and is not a phytosanitary treatment. They are separate claims and conflating them causes both mould claims and border holds.
Lead times
| Service | Lead time | Minimum | Notes |
|---|---|---|---|
| Paid pallet collection, Austin metro | 48 hours, routed | 100 units or 1 trailer | Counted with your team present, signed on the dock |
| Paid collection, Texas Triangle | Weekly, fixed day | 1 trailer | Backhaul lanes to SA, Houston, DFW |
| Yard drop-off | Open during yard hours | 5 units | Saturday mornings are easiest; no appointment |
| Spot load, common footprint | 2–5 business days | 25 units delivered | Subject to grade availability |
| Standing order | Fixed day each week | By agreement | Annual pricing, ≈2 weeks of safety stock held here |
| Repair, toll (your units) | 3–7 business days | 100 units | Your asset numbers and stamps retained |
| Heat treatment & ISPM-15 stamping | Usually 24 hours | By agreement | 4–9 hour cycle; chart trace filed per lot |
| New build to spec | 5–10 business days | 100 units | Tolerance ±⅛" achievable |
| Cut-down to custom footprint | 3–6 business days | 50 units | Tolerance ±¼" |
| Custom crate, one-off | 3–8 business days | 1 | On-site crating available in Austin metro |
| Custom milling, cut list | 3–6 business days | 1 shop hour | Tolerance ±1/16" dimensional |
| Bulk mulch or bedding | 2–5 business days | 10 cu yd | 60 mile delivery radius |
| Sourcing audit | By arrangement | 500 units/week | Half a day on site; free above that volume |
| Same-day, Austin metro | Same day | By arrangement | Premium rate, when a line is about to stop |
All times Central. Lead times are commitments we take seriously and not guarantees against weather, road closure or equipment failure.
Commercial appendix
The arithmetic behind a pallet decision, the sector matrix, and the four decision trees we walk customers through.
Cost per trip, worked
| Line | One-way recycled | Repaired recycled, 6 trips | New build, 10 trips | Pooled rental |
|---|---|---|---|---|
| Unit cost delivered | $11.50 | $13.00 | $26.00 | $0 capital |
| Useful trips | 1 | 6 | 10 | per trip |
| Cost of the asset per trip | $11.50 | $2.17 | $2.60 | $5.50 issue fee |
| Freight per trip | $1.80 | $1.80 | $1.80 | included |
| Repair share per trip | — | $0.55 | $0.42 | included |
| Handling labour per trip | $0.90 | $0.90 | $0.90 | $0.90 |
| Loss / non-return per trip | n/a | $0.30 | $0.75 | $2.10–6.00 |
| Disposal | $0.40 | credit | credit | n/a |
| Cost per trip | $14.60 | $5.72 | $6.47 | $8.50–12.40 |
A worked illustration using mid-2020s Central Texas figures, not a quotation. The decisive line is loss: pooling and new builds are both excellent until units stop coming back, and the pooled range above reflects return rates between 90% and 55%. Model your own return rate before you model anything else.
Building your own model
- Delivered unit cost. Not the quoted unit price — add the freight to your dock.
- Useful trips per unit. Ask your supplier, then halve it until you have your own data.
- Return rate. Of units leaving your building, how many come back into your control? Operations assume 80% and discover 45%.
- Repair cost and repair rate. What share needs work per cycle, and at what price.
- Handling labour. Count touches, multiply by minutes, multiply by fully loaded labour cost.
- Disposal or credit. A yard that pays for broken units turns this line from a cost into revenue.
- Failure cost. Rare, expensive, and the reason cheapest-per-unit is not cheapest-per-trip.
- Then divide. Total cost ÷ useful trips. That is the only number worth negotiating on.
What each sector actually needs
| Sector | Usual footprint | Usual grade | The spec that matters most | Failure that costs most |
|---|---|---|---|---|
| Grocery & retail distribution | 48×40 | A / #1 | Deck gap and appearance | Rejection at a retail DC door |
| Food manufacturing | 48×40 | A / #1 | Moisture ≤15% | Mould transfer to packaging |
| Beverage | 48×40, 40×48 | A / B | Racked rating, deck strength | Rack deflection under dense load |
| Construction & building supply | 48×40, 48×48 | B / #2 | Racked rating, outdoor creep | Rack incident under bagged goods |
| Agriculture & produce | 48×40, bins | B, HT only | Ventilation, treatment segregation | Chemically treated wood near feed |
| Chemicals & paint | 42×42, 48×48 | A / B | Square footprint, drum stability | Drum shift in transit |
| Automotive | 48×45, custom | A / new | Tolerance, returnable design | Line stoppage |
| Pharmaceutical | 48×40, new | New / plastic | Documented custody, cleanability | Audit failure |
| 3PL & warehousing | Mixed | Per client | Asset tracking, written standards | Unbilled pallet leakage |
| Manufacturing & electronics | Custom, 48×40 | A / #1 / new | Squareness, entry height | ASRS rejection, jammed jack |
| Export & freight forwarding | Destination's | B + HT | ISPM-15 truth | Container held on a quay |
| Hospitality & events | 48×40, half | B / #2 | Collection frequency | Service corridor blocked |
Decision trees
Recycled, new, or cut down?
- Does a standard footprint fit your carton pattern with no overhang? If yes → recycled.
- If no: does an oversized footprint exist in reclaim that can be sawn to your size? If yes → cut down, at roughly a third of a new build.
- Does anything automated read the pallet's geometry, or do you need ±⅛"? If yes → new build.
- Is the racked load above about 2,500 lb? If yes → new build, mixed hardwood, block construction.
- Is a documented chain of custody for the wood itself part of an audit? If yes → new build.
- Otherwise → recycled, at the lowest grade that survives the journey.
Buy, own a fleet, or rent from a pool?
- Measure your return rate first. Everything below depends on it and nothing else matters as much.
- Below 50% return → your pallet is packaging. Buy recycled one-way and price it as consumable.
- 50–85% return → the messy middle. Buy recycled on a managed programme, with collection on the same movements.
- Above 85% return, with controlled geography → own a fleet and use toll repair to keep your asset numbers.
- Mandated by a retailer's pool, or shipping to hundreds of points needing identical quality → pool rental, and budget for loss charges.
- Wash-down hygiene is a process requirement → plastic, and we will point you to somebody who sells it.
Stringer or block?
- Does it go into racking where you cannot guarantee orientation? Block.
- Do powered jacks enter from the 48" side? Block.
- Automated storage and retrieval? Block, or new.
- Heavy load sitting for months? Block, mixed hardwood.
- European or ISO-metric footprint? Block by specification, not preference.
- Otherwise → stringer: lighter, cheaper, deeper in recycled stock and easier to repair.
Repair or replace?
- Crack crossing a notch, or more than a third of a stringer's depth? Plug, or condemn.
- Three or more stringer repairs already present? Condemn — that is a pallet held together by optimism.
- Four or more deck boards gone on a low-value footprint? Harvest the boards.
- Above 24% moisture? Dry stack first, then decide.
- Contaminated with oil, chemicals or paint? Out of every reuse stream, permanently.
- Otherwise → repair and re-nail fully, at roughly a third of replacement cost.