Most fashion retailers understand the pressure of predicting total product demand, but high-level numbers only tell half the story. To truly optimize your stock, you must also know which specific sizes will sell, where they are needed, and when the demand will peak. Size curve forecasting is the process of predicting how customer demand for a product is distributed across its size range.
By mastering size-level predictions, apparel retailers can significantly reduce broken size runs, minimize missed sales, and prevent markdowns associated with excess inventory. This article explains how size curve forecasting works, why it matters for your bottom line, and how you can use these insights to make smarter, data-driven inventory decisions.
Key Takeaways
- Size curve forecasting allows fashion retailers to predict demand at the variant level, ensuring the right size mix is always available to meet customer demand.
- Accurate size curves lead to better buying, more precise localized allocation, and optimized replenishment, which protects your retail profit margins.
- Success depends on analyzing clean historical sales data, size distribution patterns, and location-specific data while adjusting for stockouts and seasonal shifts.
- Inventory Planner helps retailers leverage advanced forecasting and replenishment planning to make informed decisions across all products, variants, and sales channels.
What Is Size Curve Forecasting?
Size curve forecasting is the analytical process of estimating the expected demand percentage for each size within a product or category. While a standard demand forecast tells you how many total units of a style you might sell, the size curve determines the specific size distribution of those units. For example, a retailer planning for apparel, footwear, or accessories uses these curves to ensure they don’t over-invest in fringe sizes while running out of popular sizes.
Consider a simple example: if you expect to sell 1,000 units of a new t-shirt, a generic buy could leave you with too many small sizes and no large sizes. However, if your size curve analysis predicts a distribution of 10% Small, 30% Medium, 35% Large, and 25% Extra Large (XL), you can guide your purchasing with precision. This distinction is vital because customers do not buy “overall inventory” in a general sense; they buy the exact size that fits them. Unavailable sizes result in lost sales and frustrated shoppers, even when total stock levels appear perfectly healthy.
Why Size Curve Forecasting Matters for Fashion Inventory
Operating without structured size-level insights often leads to operational friction, unsold sizes, and missed sales. Because the fashion industry relies on specific fits, having the right style in the wrong size is effectively the same as being out of stock.
By prioritizing size curve forecasting, retailers gain several strategic advantages:
- Reduces broken size runs: You can prevent the common “Swiss cheese” inventory effect, where stock is available, but the most popular sizes are completely missing.
- Improves purchasing accuracy: Buyers can order the exact mix of sizes needed for a specific season, rather than relying on generic pre-packs or supplier guesswork.
- Supports better allocation: Size curve distribution patterns often vary by store location, region, or sales channel. Size curves let you tailor your stock to localized demand across stores.
- Minimizes slow-moving overstock: By identifying the different sizes that historically move more slowly, you can avoid tying up valuable working capital in units that will eventually require heavy markdowns.
- Protects profit margins: Better size-level planning leads to higher full price sales, reducing the need for steep end-of-season clearance events.
- Enhances customer experience: Shoppers are more likely to remain loyal to your brand when they can consistently find their preferred size in stock.
Size Curve Forecasting vs. Standard Demand Forecasting
It is helpful to view size curve forecasting and standard demand forecasting as complementary layers of the same inventory planning strategy. Standard demand forecasting typically estimates the total units you expect to sell for a product, category, or sales channel over a set period. It answers the basic question: “How many units do we need in total?”
Size curve forecasting, on the other hand, takes that total forecast value and distributes it across individual sizes or variants. For instance, if you forecast that a specific dress will sell 500 total units, you apply an optimal size curve to determine how many of those units should be XS, S, M, L, and XL. Without this second layer of variant planning, you risk buying the right product but in the wrong size mix, leading to simultaneous stockouts and overstocks. Combining both methods ensures your planning is broad enough to cover overall demand and granular enough to satisfy every customer.
How the Size Curve Forecasting Process Works
Successful size curve forecasting is an ongoing planning process that should evolve alongside your actual demand, sales data, and customer behavior. Retailers typically find that they need separate curves for different product attributes, categories, fits, or even specific sales channels to maintain accuracy.
1. Gather sales and inventory data by size
Begin by collecting historical sales data. You need a clear view of units sold, current inventory levels, sell-through rates, and returns for every individual size. This data provides the foundation for understanding your optimal size curve.
2. Clean and normalize the data
Raw sales data can be misleading. You must adjust for periods when the same sizes were out of stock, as a lack of sales might simply mean the sizes weren’t available to buy. Consider unusual promotional periods, inconsistent size naming across brands, and discontinued products.
3. Identify size-level demand patterns
Once the relevant data is clean, calculate the percentage of total demand each size represents within a specific category or location. This creates a size profile that reflects true demand rather than just what was sold based on historical, unadjusted numbers.
4. Apply the size curve to future demand forecasts
Take your overall product demand forecast and multiply the total units by the percentages in your size curve. This translates a high-level forecast into a detailed requirement for each variant SKU.
5. Use the forecast to guide buying and allocation
Use these calculated numbers to inform your purchase orders, initial store allocations, and replenishment schedules. This ensures your purchasing decisions are backed by data rather than generic pack ratios or pre-packs.
6. Compare forecasts to actual results
Review your performance at the end of each season. Compare your predicted size distribution against actual sales and stockouts to refine your smarter size curve for the next planning cycle.
How to Calculate Size Curves
Calculating an accurate size curve involves more than simply dividing the sales of one size by your total sales. While that provides a basic baseline, true optimization requires normalizing your sales data to account for periods when specific sizes were out of stock. If you do not adjust for these stockouts, you will calculate a skewed curve that perpetuates inventory imbalances.
Step 1: Calculate the Raw Sales Percentage
First, determine the baseline contribution of each size during a specific period. Divide the units sold for a specific size by the total units sold across the entire size run:
Raw Size % = (Units Sold of Specific Size / Total Units Sold) X 100
Step 2: Adjust for Stockouts and Lost Sales
To prevent a historically sold-out size from looking unpopular, calculate the adjusted demand. Identify the average daily sales rate for each size only during the days it was actually in stock, then multiply it by the total active selling days of the season:
Adjusted Demand of Specific Size = Average Daily Sales Rate of Specific Size X Total Active Selling Days
Step 3: Define the Optimal Size Curve
Finally, calculate the new, normalized percentage for each size based on the adjusted demand. This gives you your optimal size curve, ready to be applied to future high-level forecasts:
Optimal Size Curve % for Specific Size = (Adjusted Demand of Specific Size / Total Adjusted Demand of All Sizes) X 100
Example of Size Curve Forecasting in Fashion Retail
Imagine a retailer planning the inventory for a new seasonal women’s sweater. Based on historical sales of similar items, the team expects to sell 2,000 units over the season.
Instead of making the common human error of ordering 400 units per size, the retailer applies their historical size curve to determine the optimal size run.
The resulting buying plan is structured as follows:
| Size | Size Curve % | Calculated Units to Buy |
| XS | 8% | 160 units |
| S | 18% | 360 units |
| M | 32% | 640 units |
| L | 27% | 540 units |
| XL | 15% | 300 units |
| Total | 100% | 2,000 units |
By using this distribution, the retailer avoids an imbalanced buy. This data-driven approach ensures they have more stock in the popular Medium and Large sizes while maintaining leaner inventory for XS. While this is a simplified example, professional planners often create separate curves for different store clusters or e-commerce channels to account for regional fit and consumer preferences.
Common Size Curve Forecasting Challenges
Even with a structured process, several hurdles can distort your size-level projections. Recognizing these challenges is the first step toward building a more reliable strategy.
Incomplete or misleading sales data
If a specific size is out of stock for three weeks during a peak period, your sales history will register artificially low demand. This “lost sales” data must be accounted for; otherwise, your next forecast will tell you to buy less of a popular size that actually has high demand.
Inconsistent sizing across products
Retailers often deal with fit differences even within the same brand. A “Medium” in a slim-fit shirt may have a different demand profile than a “Medium” in a relaxed-fit style. Applying one universal curve across different cuts can lead to inventory imbalances.
New products with limited historical data
Forecasting size demand for a first-time brand launch or a completely new category is challenging. In these cases, planners often use “proxy” data from similar items or normal distribution models until actual sales history is established.
Regional and channel differences
Customer preferences often shift by location. For example, a brick-and-mortar store in one city may sell more of the smaller sizes, while your e-commerce channel sees higher demand for XL. Relying on a single global curve ignores these vital local patterns.
Overreliance on static size ratios
Suppliers often ship inventory in “packs” (or pre-packs), which are pre-assembled bundles containing a fixed, predetermined ratio of sizes (such as 1-2-2-1). While standard pre-packs simplify manufacturer logistics, blindly relying on them without pack optimization, backed by data analytics, creates severe mismatches. If your actual demand doesn’t match these ratios, you will constantly struggle with unsold sizes and stockouts.
Difficulty acting quickly on size-level insights
A precise size forecast is only valuable if your team can translate it into purchase orders. If your replenishment process is slow or manual, you cannot react quickly enough to save a broken size run or take advantage of shifting trends.
Best Practices for More Accurate Size Curve Forecasting
To maintain a competitive edge, your forecasting should be dynamic and integrated into your daily operations. Consider these best practices to improve the precision of your size-level planning.
Use size-level sales and inventory data
Ensure you are tracking both what was sold and what was available. True demand is a combination of actual sales and the potential sales you could have made if the right sizes were in stock during key selling periods.
Adjust for stockouts and broken size runs
Always “normalize” your data by estimating missed sales during out-of-stock periods. This prevents you from under-ordering popular sizes in the future simply because they weren’t on the shelf to be sold last time.
Segment size curves by product type
Avoid the trap of a one-size-fits-all curve. Create separate profiles for denim, dresses, outerwear, and footwear. The size distribution for a pair of leggings is rarely the same as the distribution for a structured blazer.
Review size curves by channel and location
Analyze how size demand differs between your physical stores, online marketplace, and wholesale channels. Tailoring your allocation to these specific demand signals ensures your stock is always in the right place.
Reforecast throughout the season
Treat your forecasts as living documents. Review your size-level performance weekly during high-volume periods or local events to catch emerging trends or sudden shifts in customer preferences.
Connect forecasts to replenishment and purchasing decisions
Data is only valuable if it triggers action. Ensure your size curve insights directly inform your replenishment recommendations, transfers between locations, and markdown strategies to optimize overall inventory.
How Inventory Planner Provides a Smarter Approach to Size-Level Inventory Planning
For modern fashion brands, managing thousands of SKUs across multiple channels using manual spreadsheets is a recipe for error. To scale effectively, you need an intelligent, centralized system that bridges the gap between high-level demand and variant-level execution. This is where Inventory Planner becomes a vital part of your tech stack.
The platform is designed to handle the complexities of fashion retail by providing:
- Variant-level demand forecasting: Plan demand patterns at the individual size, variant, and color level based on robust historical sales data.
- Automated replenishment recommendations: Receive precise buying suggestions that tell you exactly what sizes to buy and when to buy them to meet future demand.
- Omnichannel visibility: Use advanced reporting and analytics to track stock performance, sell-through rates, and levels across all connected channels and locations in real time.
- Centralized data integrations: Connect your e-commerce platforms and POS systems to ensure you are planning with a single source of truth and minimizing human error.
- Streamlined purchasing workflows: Move beyond spreadsheets and automatically generate purchase orders based on projected demand, lead times, and supplier constraints.
Inventory Planner’s advanced reporting, variant-level analytics, and automated replenishment capabilities provide the data you need to make smarter size-level inventory decisions.
Turn Size Curve Forecasting Into Better Buying Decisions
Success in the fashion world requires planning at the level where your customers actually buy: the individual size and variant. Size curve forecasting empowers your business to move away from reactive, panic-driven ordering and toward a proactive strategy that protects your cash flow and brand reputation.
The real value of this approach comes from combining accurate demand forecasts with reliable, size-level data and automated planning workflows. When you align your inventory with your audience’s specific needs, you eliminate the guesswork that leads to costly imbalances. This ensures your brand remains agile, responsive, and consistently profitable across every collection.
If you are ready to replace manual guesswork with clear, actionable insights that scale with your brand, book a demo with Inventory Planner today and see how easy accurate variant-level planning can be.
Frequently Asked Questions
What is size curve forecasting?
Size curve forecasting is the process of predicting the percentage of demand for each individual size within a product’s size run. It uses historical sales and inventory performance data to ensure retailers buy and allocate the right size mix to meet demand.
Why is size curve forecasting important?
It is essential for preventing broken size runs and reducing stockouts in popular sizes while minimizing overstock in slow-moving ones. Accurate size-level planning improves sell-through rates, protects profit margins, and ensures a better overall experience for shoppers.
How do you calculate a size curve?
To calculate a size curve, you typically determine the percentage of total sales or demand that each size represents over a specific period. You then apply these percentages to your future total demand forecasts, making adjustments for past stockouts, seasonality, and regional differences.