Why does small size or weight drift matter so much on FBA?
Because FBA fees are tiered. Small physical changes can create non-small cost changes.
Amazon publishes fulfillment fee structures based on size tier and shipping weight, and sellers build margin models around those thresholds. If the production unit ships larger or heavier than the approved sample, the seller may face two separate cost pressures: higher inbound logistics and a possible change in Amazon fees based on Amazon’s published measurement and fee rules. That is the part many first-time buyers miss. See Amazon Seller Central’s FBA fulfillment fee documentation and size-tier references for the official fee framework.
A 6-gram variance on its own is not proof of a margin problem. Some materials absorb humidity, some packaging vendors substitute board thickness, and some insert packs change late. But a 6-gram variance stacked with box-dimension drift, unapproved packaging edits, and no pre-shipment re-measurement is a common fee-creep pattern in seller-reported discussions and Agence Octo sourcing reviews. ([Seller-reported]; [Agence Octo methodology])
A simple worked example: if your approved sample sits close to an Amazon size-tier or shipping-weight threshold, and the final packed unit comes in slightly larger because the retail box or inserts changed, the seller may need to recalculate both the expected FBA fulfillment fee and the inbound freight assumptions against Amazon’s published fee tables. The exact impact depends on the product’s starting specs, how Amazon measures the received unit, and the current Amazon fee schedule.
The rule is simple: your profit model is only as accurate as your final packed specs.
Where does size and weight variance usually come from?
Usually not from one dramatic factory mistake. More often from stacked small changes across packaging and finishing.
Common sources reported by sellers, buyers, inspectors, and third-party inspection providers include:
- thicker polybags or cartons than the approved sample
- denser foam, inserts, manuals, or accessory packs
- substituted materials with slightly different mass
- looser assembly tolerances that change final dimensions
- retail-box redesign after the sample stage
- master-carton packing changes that alter case count or carton weight
This is why a sample order tests existence. It does not test repeatability.
Factories may pull the cleanest sample configuration for approval, then assign packaging to a different line or vendor at mass-production stage. That does not automatically mean deception. It means the approved sample and the shipped unit may not come from the same control system. This is common enough in seller-reported sourcing cases and consistent with third-party inspection scope references to plan against. ([Seller-reported]; [Named third-party]; [Agence Octo methodology])
What is the Agence Octo Pain Index for size and weight variance?
The Agence Octo Pain Index is a practical buyer-risk screen for FBA sellers deciding whether their landed-margin model is stable enough to trust before shipment. It is not an Amazon fee calculator. It is a screen for how likely packed-spec variance is to destabilize fee and freight assumptions. ([Agence Octo methodology])
| Pain Index level | What it looks like | Diagnostic field | What it suggests |
|---|---|---|---|
| Low | Unit weight, retail-pack dimensions, and master-carton specs match approved records within a pre-agreed tolerance | Final packed-unit record is complete and dated | Margin model is more likely to hold |
| Medium | One spec changed, but the supplier disclosed it early and re-measured packed goods before shipment | One documented revision with re-measurement | Margin model needs recalculation before final payment |
| High | Packed dimensions were never locked, packaging changed after sample approval, or carton specs arrive only after production | Missing or late packed-spec record | Fee-band drift risk is live |
| Severe | Supplier resists third-party measurement, changes packaging vendor late, or cannot produce consistent packed-spec records | Evidence gap remains unresolved | Hold shipment until re-verified, or reconsider the supplier if the evidence gap stays unresolved |
Watch the stack, not any single signal.
What should buyers measure before paying the balance?
Measure the thing Amazon is likely to assess on, not just the naked product.
That means three separate records:
- Unit-only dimensions and weight
Useful for engineering and QC, but incomplete for FBA economics.
- Final packed-unit dimensions and weight
This is the margin-critical record for most private-label sellers, because Amazon’s fee framework is tied to the sellable unit as received and handled.
- Master-carton dimensions, gross weight, and units per carton
This affects freight quotes, pallet planning, and warehouse handling.
If the supplier sends only product specs and avoids packed specs, the measurement set is incomplete. That does not prove a problem. It shifts the burden of proof. The weaker the spec match, the more evidence the supplier needs to show. This is where Agence Octo’s sourcing reviews are most useful: pressure-testing whether the packed-spec record is complete enough to trust before funds are released.
What practical checklist should an FBA seller use?
Practical checklist
Ask for these before balance payment:
- approved spec sheet with unit dimensions and unit weight
- approved packaging spec with final packed dimensions and packed weight
- master-carton spec sheet with carton dimensions, gross weight, net weight, and units per carton
- date-stamped measurement photos with measuring tool visible
- production-run measurement sample size stated in writing
- confirmation that packaging materials match the approved sample
- written notice of any insert, manual, accessory, or carton-board change
- third-party inspection report that includes measurement checks, not just visual QC
The 20 minutes that matter: compare the approved sample record against the final packed-unit record and the master-carton record. If those three do not agree, your fee assumptions are still soft.
What are the red flags that justify a stop?
Red flags
- supplier says “same as sample” but provides no packed-spec measurements
- box dimensions are rounded instead of measured precisely
- unit weight is shared, packed weight is missing
- packaging vendor changed after sample approval
- master-carton dimensions arrive only after goods are finished
- inspection report checks workmanship but not dimensions or weight
- MOQ pushback disappears once you ask to re-measure packed goods
- supplier resists random carton pulls for re-checking
If the supplier is the only one who can answer questions about the product specs they changed, pause release until the packed specs are independently re-verified.
How should sellers think about tolerance?
Tolerance is a commercial control, not a guess.
Do not ask whether variance is “normal.” Ask what variance was agreed, what was measured, and at what production stage. Without that, the seller is comparing one approved sample against one mass-production claim. That is not control.
For FBA products near fee thresholds, even small drift deserves escalation. Per Agence Octo's sourcing methodology, the tighter your margin and the closer your packed unit sits to an Amazon fee boundary, the less tolerance you can afford. The official threshold logic comes from Amazon’s published FBA fee documentation; the sourcing risk is whether your final packed specs still fit the assumptions you modeled against. ([Agence Octo methodology]; see Amazon Seller Central fee documentation)