For cross-border sellers listing on both Amazon US (FBA) and Mercado Libre (Fulfillment), the core inventory question is not that more stock is safer, but at what inventory level each SKU should be replenished and in what quantity, so as to balance storage fees (overstock cost), low-inventory fees and stockout losses (shortage cost), and tied-up capital (holding cost). The answer comes in three steps: first convert each SKU's average daily sales and lead time into days of supply and a reorder point, then treat the platform's storage-fee age tiers as a ceiling on holding cost, and finally use safety stock to buffer demand volatility. Take Amazon's low-inventory-level fee rule from 1 February 2027: the fee is charged only when inventory falls below 28 days of supply, which sets the floor on how little to hold; Mercado Libre charges a monthly fee on goods stored more than 60 days in age tiers, while Amazon keeps fees flat or lower for inventory turning within six months and raises them for aged stock, which sets the ceiling on how much to hold.
1. The decision: how much to hold, when to reorder, how much to order
A cross-border seller listing on both Amazon US and Mercado Libre must make three decisions for each SKU in sequence: how much to hold (target inventory level), at what inventory level to reorder (reorder point), and how much to order at a time (order quantity). These three decisions drive three costs at once: holding too much means storage fees and tied-up capital (holding cost); holding too little either triggers low-inventory fees or causes stockouts that lose sales and ranking (shortage cost); ordering too often means paying inbound freight and receiving fees every time (replenishment cost). This article uses the two platforms' official rules and standard inventory formulas to quantify these three costs and give a reproducible reorder point and a hold-or-clear judgement.
2. Mechanism: how inventory turnover, days of supply and age tiers drive holding cost
Inventory turnover is an efficiency measure of how fast inventory sells, given by inventory turnover = cost of goods sold ÷ average inventory, where average inventory is the average of opening and closing inventory. It connects directly to two operating facts: the faster the turnover, the more times the same capital circulates per unit of time and the less of it is tied up; the slower the turnover, the longer stock sits and the higher the storage and holding costs.
Converting turnover into the more intuitive days of supply is more practical for replenishment decisions: days of supply = on-hand inventory ÷ average daily sales. For example, 280 units on hand at 10 units a day is 28 days of supply. This is exactly the threshold of Amazon's low-inventory-level fee — Amazon's announcement of 5 November 2026 states that from 1 February 2027 the fee applies only when inventory is below 28 days of supply, judged on actual shipped volume in the four weeks before and after the assessment date, to reduce misclassification of end-of-season or discontinued products.
The ceiling on holding cost comes from the platforms' age tiers. Amazon's same announcement says that from 1 February 2027 it will remove the storage utilisation surcharge and fold the aged-inventory surcharge into the monthly storage fee, moving to a single age-based charge, so that inventory turning within six months costs the same or less and aged inventory costs more. Mercado Libre charges a monthly per-unit fee on goods stored in its fulfilment centres for more than 60 days, tiered by age and size: for small items (≤25×33×48 cm and under 18 kg), the first four months are USD 0, months 4–6 cost USD 0.77 a month, months 6–12 USD 2.43, and beyond 12 months USD 4.14. This means holding cost is not linear but steps up with inventory age.
The cost of a stockout is not negligible either: Amazon's fulfilment data in the announcement shows that cutting the delivery promise from more than 4 days to 2 days lifted conversion by more than 20%, and cutting it again from 2 days to hours lifted conversion by a further 15% or more — a stockout means failing to honour the fast-fulfilment promise, potentially losing that conversion lift and dragging on subsequent ranking.
| Metric | Formula | Question it answers |
|---|---|---|
| Inventory turnover | Cost of goods sold ÷ average inventory | How fast the inventory as a whole sells |
| Days of supply | On-hand inventory ÷ average daily sales | How many days the current stock will last |
| Reorder point | Lead-time demand + safety stock | The inventory level at which to reorder |
| Safety stock | z × σ × √lead time | The buffer against demand volatility |
| Mercado Libre small-item age (per unit per month, USD) | First 4 months | 4–6 months | 6–12 months | Over 12 months |
|---|---|---|---|---|
| Long-term storage fee | 0 | 0.77 | 2.43 | 4.14 |
3. Reproducible analysis: reorder points and safety stock in two scenarios
Both scenarios below are explicitly labelled hypothetical examples, not data from any real store. Formula basis: lead-time demand = average daily sales × lead-time days; safety stock = z × standard deviation of daily sales × √lead time (fixed lead time, normally distributed demand, z the quantile for the service level); reorder point = lead-time demand + safety stock.
Scenario A: Amazon FBA fast-moving SKU, reorder point 159 units (hypothetical example)
Hypothetical example: average daily sales 10 units, lead time 14 days, standard deviation of daily sales σ=3, service level 95% (z=1.65), target days of supply 28 (above Amazon's low-inventory fee threshold).
Lead-time demand = 10 × 14 = 140 units; safety stock = 1.65 × 3 × √14 = 1.65 × 3 × 3.74 ≈ 18.5 units, rounded to 19; reorder point = 140 + 19 = 159 units. Replenish to 28 days of supply, that is 28 × 10 = 280 units.
Decision: when on-hand inventory falls to 159 units, trigger a replenishment and top up to 280 units. If the standard deviation of daily sales rises from 3 to 5, safety stock scales to about 31 units under the same formula and the reorder point rises to about 171 — the more volatile the demand, the more both safety stock and reorder point must rise, otherwise the same 159-unit reorder point corresponds to a higher probability of stockout.
Scenario B: Mercado Libre slow-moving SKU, stock beyond four months starts climbing the storage-fee ladder (hypothetical example)
Hypothetical example: small item (≤25×33×48 cm, under 18 kg), average daily sales 1 unit, gross margin USD 8 per unit.
If 180 units (about six months of sales) are stocked at once, about 60 units remain after month 4, entering the 4–6 month band at USD 0.77 per unit per month, then rising to USD 2.43 in months 6–12 and USD 4.14 beyond 12 months. Dividing the monthly storage fee by the USD 8 gross margin: USD 0.77 is about 9.6%, USD 2.43 about 30.4%, USD 4.14 about 51.8%.
Decision: for slow-moving SKUs, do not hold more than about four months of sales, because every extra month of holding climbs the storage-fee ladder; if a batch is not expected to sell out within 4 months, cut the purchase quantity or clear and remove it early, rather than keep paying higher age-tier storage fees. Holding more only pays off when the bulk-purchase discount is large enough to cover the storage fee for the 4–6 month band.
4. Counter-examples and alternative readings: high turnover is not always optimal, and a low-inventory fee is not always a replenishment signal
One scenario that overturns 'the faster the turnover the better': chasing high turnover by cutting order quantities too small and raising order frequency too high means paying inbound freight and receiving fees on every replenishment, so the per-unit replenishment cost rises and frequent stockouts may even lose sales. Turnover itself does not distinguish selling fast from holding little; only by looking at the stockout rate and per-unit replenishment cost together can you tell whether high turnover is efficiency or risk-taking.
Conversely, a triggered low-inventory fee does not necessarily mean you should reorder. Amazon's announcement states that one purpose of the new rule is to reduce misclassification of end-of-season and discontinued products — for an item being cleared and about to be discontinued, inventory below 28 days of supply is normal and reasonable, and replenishing would only push it into an older, more expensive age band. The way to tell: look at whether the SKU's future demand is falling passively (seasonality or discontinuation) or is an active mismatch (a reorder point set too low, causing persistent stockouts); clear the former, raise the reorder point for the latter.
5. Action plan: which fields to collect, what to use as a baseline, when to stop
- Fields to collect: for each SKU, average daily sales, standard deviation of daily sales, lead time (order to receipt), on-hand inventory, unit purchase cost, unit gross margin, monthly storage fee per unit (by age tier), and the inbound freight and receiving fee per replenishment.
- Set a baseline: record current days of supply, inventory turnover, and the number and duration of stockouts over the past 90 days, as the before-and-after comparison.
- Observation window: at least one full lead time plus one replenishment cycle (for example a 14-day lead time plus a 7-day order interval, 21 days in total), watching the stockout rate, the age distribution and the change in storage fees.
- Stop conditions: when a SKU's monthly storage fee exceeds its monthly gross profit (for example, a slow-moving item entering the 6–12 month band at USD 2.43 a month, more than 30% of gross margin), stop replenishing and clear or remove it; when the stockout rate stays above baseline, raise the reorder point and safety stock.
- Fill in data: once you have your own back-end storage-fee invoices and demand distribution per SKU, replace this article's assumed standard deviation and gross margin and recalculate.
6. Limitations and uncertainty
What this article can verify word for word is the 28-day supply threshold and the age-based fee structure in Amazon's announcement, the small-item age-tier monthly fees on Mercado Libre's official pricing page, and the standard inventory-turnover formula; what it cannot derive is Amazon's specific USD rates for age-based billing in 2027, your account's real demand distribution, and the optimal order quantity for any particular SKU.
Three unknowns are most likely to change the conclusion: first, Amazon's specific rate card for age-based billing in 2027; second, demand is not strictly normal and lead times also vary, so a fixed-lead-time safety-stock formula understates the true stockout probability; third, the Mercado Libre storage-fee table is a USD snapshot at the time of access and may change by site and exchange rate.
What evidence would update the judgement: if Amazon's specific 2027 monthly storage rates by age become readable, the 'within six months' assumption in Scenario A can be replaced with an exact USD holding cost; if you can obtain your own SKU's actual demand and lead-time distributions, the √lead time approximation can be replaced with a safety-stock formula in which both demand and lead time are random. This article discusses a general inventory analysis method under the platforms' public rules and does not imply that Nuozhou Digital Intelligence provides inventory or warehousing back-end data for any platform.
Method and notes
Data basis for this article: the inventory turnover formula, inventory turnover = cost of goods sold ÷ average inventory, and the explanation that faster turnover lowers storage and holding costs are taken from Corporate Finance Institute's Inventory Turnover; the 28-day supply threshold, the age-based fee structure and the conversion lift from faster delivery are taken from Amazon's official announcement of 5 November 2026; the monthly fee table for small items stored more than 60 days, charged in age tiers (USD 0 / 0.77 / 2.43 / 4.14), is taken from Mercado Libre's official global-selling rate page, all accessed 2026-11-19. The sales volumes, standard deviations, gross margins and stock quantities in the two scenarios are all explicitly labelled hypothetical examples, not statistics from any platform or store.
The increment here is to bring an abstract efficiency metric like inventory turnover down onto two platforms' concrete age tiers: Amazon's 28 days of supply as the floor against holding too little, Mercado Libre's monthly fee that steps up after 60 days as the ceiling against holding too much, plus the reproducible formulas reorder point = lead-time demand + safety stock and safety stock = z × σ × √lead time.
To put storage fees into a full profit account, see Cross-Border E-Commerce Profit Calculation: The Five-Step Subtraction from GMV to Net Profit; for a full breakdown of Amazon's 2027 storage and low-inventory fee changes, see 2027 Amazon FBA Fee Changes; for Mercado Libre rates and Latin American site selection, see Mercado Libre Data Analysis.
This article was assembled with the help of an automated research workflow and is based on the public sources listed at the end; it has not been manually reviewed.
References
- 1. Amazon Selling Partner Services, Update to US Referral and Fulfillment by Amazon fees for 2027 (primary source: the 28-day supply threshold, the age-based fee structure and the delivery-conversion data): read the original, accessed 2026-11-19
- 2. Mercado Libre Global Selling, Pricing and Fee Structure (primary source: the long-term storage fee table for goods stored more than 60 days, tiered by age): read the original, accessed 2026-11-19
- 3. Corporate Finance Institute, Inventory Turnover (methodology source: the inventory turnover formula and its explanation): read the original, accessed 2026-11-19