Inventory Forecasting for Ecommerce Brands

Why Forecasting Matters

Too much stock ties up cash and fills expensive warehouse space. Too little means stockouts, lost sales, and disappointed customers. Accurate forecasting sits at the heart of profitable ecommerce operations.

Forecasting Methods

1. Historical Sales Analysis

The simplest approach: use past sales data to predict future demand.

Look at the same period last year
Adjust for growth rate
Account for one-off events (viral products, PR coverage)

Best for: Established products with 12+ months of sales history.

2. Moving Average

Calculate the average sales over a rolling period (e.g., last 4 weeks) to smooth out daily fluctuations.

Simple moving average: Equal weight to each period
Weighted moving average: Recent periods count more
Exponential smoothing: Mathematically weighted towards recent data

Best for: Stable-demand products without strong seasonality.

3. Seasonal Decomposition

Break sales data into components:

Trend: Overall upward or downward movement
Seasonality: Predictable recurring patterns (Christmas, summer, etc.)
Noise: Random variation

Best for: Products with clear seasonal patterns.

4. Qualitative Forecasting

When historical data doesn't exist (new products, new markets):

Marketing team input on planned promotions
Sales team feedback on pipeline and leads
Industry trends and competitor analysis
Supplier lead time considerations

Building a Practical Forecast

Step 1: Gather Data

Last 12–24 months of sales data by SKU
Marketing calendar (planned promotions, launches, PR)
Supplier lead times
Current stock levels

Step 2: Segment Your SKUs

Segment	Characteristics	Forecasting Approach
Steady sellers	Consistent demand	Moving average
Seasonal	Predictable peaks	Seasonal decomposition
Trending	Growing or declining	Trend + growth rate
New products	No history	Qualitative + analogues
Long tail	Infrequent, low volume	Min/max stock rules

Step 3: Calculate Reorder Points

Reorder Point = (Average daily sales × Lead time in days) + Safety stock

Example: A product sells 20 units/day with a 30-day lead time and 10 days of safety stock:

Reorder Point = (20 × 30) + (20 × 10) = 600 + 200 = 800 units

When stock drops to 800 units, place the next order.

Common Mistakes

Using only averages — averages hide variability; a product selling 100 one week and 0 the next averages 50 but needs different planning
Ignoring lead times — forecasting demand is useless if you don't account for how long replenishment takes
Over-reacting to short-term trends — one viral TikTok doesn't mean permanently higher demand
Not communicating with your 3PL — your fulfilment provider needs your forecast to plan space and labour
Forecasting at the wrong level — forecast by SKU and channel, not just total units

Tools and Technology

Tool	Cost	Best For
Spreadsheets	Free	Small ranges (<100 SKUs)
Inventory Planner	£100–£500/month	Shopify/ecommerce brands
Brightpearl	£300–£800/month	Multi-channel retailers
Custom WMS reporting	Varies	Larger operations

Sharing Forecasts with Your Warehouse

Your warehousing provider should receive:

Monthly demand forecasts — for labour and space planning
Inbound delivery schedule — what's arriving and when
Promotional calendar — upcoming spikes to prepare for
New product launches — new SKUs to set up

At The Fulfilment Centre, we work closely with clients on demand planning to ensure we have the space, stock, and resources ready when you need them.

Need help with inventory planning? Contact us to discuss how we support demand forecasting and stock management.