Cycle Service Level vs Fill Rate: The Core Difference
The cycle service level vs fill rate distinction is the most misunderstood concept in inventory planning, and it leads to expensive mistakes when sellers conflate the two. Both numbers are commonly called "service level," but they measure different things.
Cycle service level (CSL) is the probability that you do not stock out at all during a single replenishment cycle. It is binary per cycle: either you ran out of any units or you did not. A 95% CSL means that across many cycles, you avoid stockouts in 95% of them.
Fill rate measures the percentage of unit demand you actually fulfill. A 95% fill rate means that across all the units customers tried to buy, you shipped 95% of them. The remaining 5% were lost or backordered.
The cycle service level vs fill rate gap matters because the same safety stock policy produces different scores on each metric. A 95% CSL typically corresponds to a 98 to 99% fill rate, because most failed cycles only stock out for the last day or two of a 30 to 60 day cycle. The unit-weighted view paints a much more accurate picture of customer impact.
Cycle Service Level vs Fill Rate Formulas
The cycle service level formula uses the standard normal distribution:
CSL = P(demand during lead time ≤ reorder point)
= Φ(Z)Where Φ is the cumulative standard normal function and Z is your Z-score service factor. A Z of 1.65 produces CSL = 95%.
The fill rate formula is more involved because it accounts for partial fulfillment:
Fill Rate = 1 - (Expected Shortage per Cycle / Expected Demand per Cycle)
= 1 - (E(z) × σ_dLT) / DWhere E(z) is the standard normal loss function for your Z, σ_dLT is the standard deviation of demand during lead time, and D is expected demand per cycle. The loss function E(z) for common Z values:
| Z (Service Factor) | CSL % | E(z) Loss Function | Resulting Fill Rate (typical) |
|---|---|---|---|
| 1.28 | 90.0% | 0.0475 | ~97% |
| 1.65 | 95.0% | 0.0206 | ~98.5% |
| 1.96 | 97.5% | 0.0094 | ~99.3% |
| 2.33 | 99.0% | 0.0033 | ~99.75% |
The takeaway: a 95% CSL gets you to roughly 98.5% fill rate. The cycle service level vs fill rate gap is consistently 3 to 6 points.
Worked Example: Cycle Service Level vs Fill Rate on the Same SKU
You sell a $48 ASP home goods product. Average daily demand is 32 units. Lead time is 60 days. Standard deviation of lead time demand is σ_dLT = 95 units. You set a 95% cycle service level target (Z = 1.65).
Safety stock = 1.65 × 95 = 157 units.
Expected demand per cycle = 32 × 60 = 1,920 units.
Expected shortage per cycle = E(1.65) × σ_dLT = 0.0206 × 95 = 1.96 units.
Fill rate = 1 - (1.96 / 1,920) = 99.9%.
Wait, that's higher than the table suggests. The table assumes σ_dLT/D is at a typical ratio. On this SKU, σ_dLT is small relative to expected demand (95 / 1,920 = 0.05), so fill rate is very high.
Now compare a noisier SKU: σ_dLT = 280 units on a 1,920-unit cycle.
Expected shortage = 0.0206 × 280 = 5.77 units.
Fill rate = 1 - (5.77 / 1,920) = 99.7%.
Same 95% CSL target, but the cycle service level vs fill rate translation depends on how noisy the SKU is. High-variability SKUs need a bigger lift over baseline CSL to hit the same fill rate.
FBA-Specific Context: Why Fill Rate Wins for FBA
FBA economics push you toward fill rate as the primary metric for three reasons.
1. Lost Buy Box time is unit-weighted. When you stock out, Amazon shifts the Buy Box to a competitor or unsuppresses your listing entirely. The financial damage scales with units of demand missed, not whether the cycle technically failed. Fill rate captures this; cycle service level does not.
2. Subscribe & Save is unit-driven. If your S&S customers cannot get their order on the scheduled date, Amazon either delays or cancels. Each cancelled S&S order is a unit of lost revenue and a permanent customer loss. Fill rate maps directly to this damage.
3. Out-of-stock rate in FBA reports is closer to fill rate. Amazon Seller Central does not surface cycle service level. The reports you actually see (in-stock rate, lost sales) align with fill rate math, so optimizing for cycle service level vs fill rate when fill rate is what gets reported wastes effort.
To translate a target fill rate into the actual buffer units, you can calculate safety stock with our free tool.
Common Mistakes
1. Treating CSL and fill rate as the same number. Most safety stock formulas in textbooks deliver cycle service level. Sellers often plug a fill rate target (98%) into a CSL-shaped formula and end up under-buffered by 30 to 50%.
2. Setting a 99% fill rate on every SKU. The marginal cost of moving fill rate from 98% to 99% is roughly equal to moving from 90% to 98%. The last point is brutally expensive. Reserve 99%+ fill rate targets for SKUs where the contribution margin clearly justifies it.
3. Reporting one and budgeting for the other. When the leadership team asks for a "service level," confirm whether they mean cycle service level vs fill rate. Reporting 95% CSL when they expected 95% fill rate makes you look better than reality, until a quarter goes wrong and the gap shows up.
Related Terms
Frequently Asked Questions
Which is better for FBA: cycle service level or fill rate?
Fill rate is the better metric for FBA because it weighs by units. A 95% cycle service level can still mean 5% of unit demand goes unfulfilled, which on a high-velocity SKU is significant lost revenue. Fill rate directly measures the customer experience and revenue capture.
Why is fill rate always higher than cycle service level?
Cycle service level is binary per cycle: any stockout, even one unit, fails the cycle. Fill rate measures partial credit. If you stock out for the last 3 days of a 30-day cycle, cycle service level scores zero for that cycle but fill rate counts the 27 successful days. Fill rate is typically 3 to 6 percentage points higher than cycle service level for the same safety stock.
What target fill rate should I set?
For most FBA sellers, a 98% fill rate target on hero SKUs and 95% on mid-tier SKUs is reasonable. Going to 99%+ requires meaningful extra safety stock and is only justified for highly competitive categories or Subscribe & Save anchor SKUs.
How do I calculate fill rate after the fact?
Fill Rate = Units Shipped / Units Demanded over a period. For FBA, units demanded is harder to measure because lost sales during stockouts do not show up in your data. Estimate by extrapolating average daily demand into the stockout period, then subtract that estimate from total potential demand.
Does Amazon care about my fill rate?
Amazon does not directly score fill rate, but it does score the upstream effects: out-of-stock rate, lost Buy Box time, and IPI score components. A low fill rate indirectly hurts your IPI through excess and stranded inventory penalties when sellers overcorrect with extra buffer.