How Amazon Fulfillment Centers Work: A Practical Overview
20 December 2025
Amazon FBA vs FBM vs Vendor Central: Which Model Fits Your Business?
20 December 2025
OUR GOAL
To provide an A-to-Z e-commerce logistics solution that would complete Amazon fulfillment network in the European Union.
If you manage an e-commerce inventory, you likely know your "A-items" by heart. They are your revenue drivers, the 20% of SKUs that generate 80% of your turnover. You watch them like a hawk. But here is the uncomfortable truth: knowing how much revenue an item generates tells you absolutely nothing about how difficult it is to manage.
This is where the standard ABC analysis fails. It is one-dimensional. It assumes that a product bringing in €10,000 a month is always an "A" priority, regardless of whether that €10,000 comes from steady daily sales or a single, unpredictable bulk order.
For logistics managers and supply chain directors, volume is vanity, but predictability is sanity. To bridge the gap between revenue potential and operational reality, you must introduce a second dimension to your matrix: volatility.
Welcome to the XYZ Analysis.
Beyond Pareto: Why revenue isn't enough
The classic ABC analysis (based on the Pareto Principle) classifies inventory based on consumption value. It answers the question: "Which items make us the most money?"
- A-items: High value (top 70-80% of revenue).
- B-items: Moderate value (next 15-20%).
- C-items: Low value (bottom 5-10%).
However, in the fast-paced world of e-commerce, where trends shift overnight due to TikTok vitality or seasonal spikes, relying solely on value is dangerous. It leads to a common phenomenon: The "A-item" stockout.
You might treat two A-items identically because they have the same annual revenue. Yet, Item 1 sells 10 units every day like clockwork, while Item 2 sells nothing for three weeks and then 300 units in two days. If you use the same restocking logic for both, you will inevitably overstock Item 1 (tying up capital) and stock out of Item 2 (losing sales).
XYZ analysis solves this by classifying items based on demand fluctuation.
Decoding the XYZ classifications
While ABC looks at value, XYZ looks at the Coefficient of Variation (CV) over a specific period. It answers the question: "How hard is it to predict future demand for this item?"
X-items: Steady sellers
These are characterized by steady turnover and minimal fluctuation. Demand is constant, reliable, and easy to forecast.
- Characteristics: Low volatility, high forecast accuracy.
- Logistics implication: You can operate these with lean inventories and "Just-in-Time" (JIT) replenishment methods. Automated reordering works perfectly here.
- Example: Basic white t-shirts, printer paper, or standard charging cables.
Y-items: Fluctuating movers
These items show stronger variations in demand. The fluctuation might be due to seasonality (e.g., swimwear), marketing campaigns, or product lifecycles.
- Characteristics: Moderate volatility, forecastable but with a margin of error.
- Logistics implication: These require higher safety stocks than X-items. Manual intervention is often needed to adjust forecasts based on known external factors (e.g., an upcoming Black Friday sale).
- Example: Winter coats, garden furniture, or trend-driven electronics.
Z-items: Erratic unpredictables
Demand for these items is sporadic. They might see zero sales for weeks, followed by a sharp spike, or vice versa.
- Characteristics: High volatility, extremely difficult to forecast.
- Logistics implication: Automation fails here. You need either high safety stocks (which is expensive) or a "Make-to-Order" strategy. These are the biggest headache for warehouse managers.
- Example: Spare parts for discontinued models, niche luxury items, or new product launches with no historical data.
Math behind the madness: Calculating coefficient of variation
To implement this professionally, you cannot rely on "gut feeling." You need to calculate the Coefficient of Variation (CV) for each SKU. This removes bias from the process.
The formula is:
CV = Standard Deviation of Demand
    Average Demand
How to interpret the results:
- X class: CV less than or equal to 0.5 (Variation is less than 50% of the average).
- Y class: greater than 0.5 but less than or equal to 1.0 (Variation is significant).
- Z class: CV is greater than 1.0(Variation is higher than the average demand itself).
Note: The thresholds (0.5 and 1.0) can be adjusted based on your specific industry vertical and risk appetite.
9 categories of inventory
When you overlay the XYZ volatility axis onto the ABC value axis, you get a 9-box matrix. This is the "Holy Grail" of inventory optimization. Each box requires a distinct logistics strategy.
The "A" row (High value)
- AX (High value, steady demand): The Gold Standard. These are your cash cows. Because demand is stable, you should aim for the lowest possible stock levels to free up working capital. Automated replenishment is safe here.
- AY (High value, fluctuating demand): The Danger Zone. High capital is tied up here, but demand swings. You need close human supervision. Use advanced forecasting tools that account for seasonality.
- AZ (High value, erratic demand): The Nightmare. These items are expensive to hold and hard to sell. Avoid stocking these locally if possible. Consider drop-shipping or centralizing stock in a single hub rather than distributing it across multiple fulfillment centers.
The "B" Row (Medium value)
- BX (Medium value, steady demand): Manage these similarly to AX but with less frequent review intervals. They are stable contributors.
- BY (Medium value, fluctuating demand): Requires a buffer. The cost of a stockout is moderate, so carrying slightly more safety stock is usually cheaper than the administrative cost of micromanaging the forecast.
- BZ (Medium value, erratic demand): Review the necessity of these items. If the holding cost exceeds the profit margin, consider delisting them or switching to an on-demand model.
The "C" Row (Low value)
- CX (Low value, steady demand): Easy to manage. Order in bulk to secure quantity discounts and reduce shipping frequency. Since the value is low, carrying 6 months of stock is often acceptable to minimize administrative hassle.
- CY (Low value, fluctuating demand): maintain higher safety stocks. The risk of obsolescence is low (usually), and the capital cost is negligible.
- CZ (Low value, erratic demand): The "Dead Stock" Candidates. These clutter your warehouse management system (WMS) and physical shelves. Unless they serve a strategic purpose (like a critical spare part), aggressive rationalization is recommended.
Strategic implications for warehousing and fulfillment
Understanding the matrix is step one. Step two is applying it to your physical operations. For a business scaling in Europe, particularly when leveraging a partner like Flex Logistique, this data dictates how your goods should be physically handled.
1. Warehouse layout optimization
Your AX and BX items are high-turnover and predictable. They should be placed in the "Golden Zone" of the warehouse—waist height, near the packing stations—to minimize picking travel time.
Conversely, Z-items (erratic) can be placed in higher racking or deeper storage zones (mezzanines), as pickers won't need to access them frequently. This slotting strategy alone can improve picking efficiency by 20-30%.
2. Safety stock calibration
The "one-size-fits-all" safety stock formula is a profit killer.
- For X-items: You can reduce safety stock significantly. If you sell exactly 100 units a week, you don't need a 50% buffer.
- For Z-items: You face a trade-off. You either accept stockouts (lowering customer service levels) or hold high safety stock (increasing storage costs). The XYZ analysis allows you to make this decision consciously rather than accidentally.
3. Purchasing strategy
- X-items: Negotiate blanket orders with suppliers with scheduled JIT deliveries. You have the data to guarantee volume.
- Z-items: Do not commit to volume. Negotiate for shorter lead times or smaller Minimum Order Quantities (MOQs), even if it means a slightly higher unit price. Flexibility is worth the premium here.
Role of automation and tech stack
In modern logistics, you cannot calculate XYZ manually in Excel once a year. It must be dynamic.
Most advanced Warehouse Management Systems (WMS) or ERPs can calculate standard deviation. However, the trick is integrating this into your reordering points.
- Automate X: Set your system to reorder AX and BX items automatically when they hit the reorder point. There is no need for a human to approve a purchase order for printer paper.
- Flag Y: Set the system to alert a procurement manager. "System suggests ordering 500 units, but volatility is detected."
- Review Z: Disable auto-reordering. Every purchase of a Z-item should require manual approval to prevent inventory bloat.
When "Z" becomes a problem: The 3PL advantage
One of the biggest challenges for growing e-commerce brands is managing the "Z" tail—the erratic sellers that take up space.
This is where a Third-Party Logistics (3PL) provider becomes a strategic asset rather than just a cost center. A robust fulfillment partner allows you to convert fixed costs into variable costs.
- Scalability: If a "Z-item" suddenly spikes (perhaps due to an influencer mention), a 3PL has the labor and space elasticity to handle the surge without you needing to rent more warehouse space permanently.
- Pooling risk: By centralizing inventory with a logistics partner, you mitigate the risk of having stock trapped in the wrong regional hub.
Furthermore, experienced logistics partners often have WMS capabilities that provide this visibility for you. Instead of building your own analytics stack, you leverage theirs to identify which SKUs are costing you money in storage fees versus which are turning a profit.
Turning volatility into opportunity
The goal of inventory management is not to eliminate volatility—that is impossible in the modern market. The goal is to domesticate it.
By adding the XYZ dimension to your ABC analysis, you move from a reactive state to a proactive one. You stop over-servicing low-risk items and stop under-estimating high-risk ones. You align your capital investment with reality.
In the end, inventory optimization is about balance. The AX items pay the bills, but effectively managing the Z items protects your margins. Start by calculating your coefficients today. You might find that your "best" products are actually your most dangerous logistical liabilities, and that is the first step toward fixing your supply chain.
