Data-Driven Buying: Using Analytics to Order the Right Sizes

For wholesale buyers and menswear retailers, few decisions have as much impact on profitability as size allocation. Order too many small sizes and you're stuck with unsellable inventory; order too few large sizes and you lose sales to stockouts. Traditional gut-feel buying leaves money on the table, while data-driven size allocation optimizes inventory investment, maximizes sell-through, and minimizes markdowns. This comprehensive guide provides a systematic approach to using analytics for smarter size buying decisions.

In this detailed analysis, we'll explore how to collect and analyze size data, interpret sales patterns, adjust for market variations, implement predictive buying models, and build a data-driven size allocation strategy that drives profitability.

The Cost of Poor Size Allocation

Understanding the financial impact justifies investment in analytics.

Direct Financial Costs

Excess Inventory (Wrong Sizes):

• Capital tied up in unsellable stock
• Storage costs for aging inventory
• Markdowns to clear (30-70% off typical)
• Opportunity cost (could have bought right sizes)
• Estimated impact: 15-25% margin erosion

Stockouts (Missing Sizes):

• Lost sales (customer buys elsewhere)
• Lost customer lifetime value
• Negative word-of-mouth
• Damaged reputation
• Estimated impact: 10-20% revenue loss

Example Scenario:

• Order 100 suits at $200 each = $20,000 investment
• Poor allocation: 40% wrong sizes
• Markdown loss: $8,000 × 50% = $4,000 lost
• Stockout loss: 20 missed sales × $600 retail = $12,000 revenue lost
• Total impact: $16,000 (80% of investment)

Operational Costs

Inefficiency:

• Time spent managing excess inventory
• Frequent reordering of stockout sizes
• Customer service handling complaints
• Staff frustration and morale impact

Essential Data to Collect

Effective analytics starts with comprehensive data collection.

Sales Data by Size

What to Track:

• Units sold per size (by SKU/style)
• Revenue per size
• Sell-through rate per size
• Time to sell per size
• Stockout frequency per size
• Return rate per size

How to Capture:

• Point-of-sale system reports
• Inventory management software
• E-commerce platform analytics
• Manual tracking if necessary

Minimum Data Period:

• Ideal: 12 months (full seasonal cycle)
• Acceptable: 6 months (half cycle)
• Minimum: 3 months (limited but useful)

Inventory Data

What to Track:

• Opening inventory by size
• Receipts (new inventory) by size
• Sales by size
• Returns by size
• Markdowns by size
• Closing inventory by size

Key Metrics:

• Inventory turnover by size
• Days of supply by size
• Sell-through percentage by size
• Weeks of supply remaining

Customer Data

What to Track:

• Size preferences by customer segment
• Repeat purchase patterns
• Size exchanges and returns
• Customer demographics (age, location)
• Purchase occasion (work, formal, casual)

Market Data

External Factors:

• Regional body type variations
• Demographic shifts in market
• Competitor size availability
• Industry size distribution benchmarks
• Seasonal variations

Analyzing Size Performance

Transform raw data into actionable insights.

Sell-Through Analysis

Formula:

• Sell-Through % = (Units Sold ÷ Units Received) × 100

Example Analysis (Dress Shirts):

• Size 15: 45 sold ÷ 50 received = 90% sell-through ✓ Good
• Size 15.5: 48 sold ÷ 50 received = 96% sell-through ✓ Excellent (consider increasing)
• Size 16: 42 sold ÷ 50 received = 84% sell-through ✓ Good
• Size 16.5: 38 sold ÷ 50 received = 76% sell-through ⚠ Acceptable
• Size 17: 25 sold ÷ 50 received = 50% sell-through ✗ Poor (reduce allocation)
• Size 17.5: 15 sold ÷ 50 received = 30% sell-through ✗ Very poor (significantly reduce)

Interpretation:

• 90%+ sell-through: Increase allocation (potential stockouts)
• 75-89% sell-through: Maintain allocation (healthy)
• 60-74% sell-through: Slight reduction (excess inventory)
• Below 60%: Significant reduction (major excess)

Stockout Analysis

What to Track:

• Frequency of stockouts per size
• Duration of stockouts
• Lost sales estimates
• Customer complaints

Example:

• Size 16 shirts: Stocked out 4 times in 6 months
• Average stockout duration: 2 weeks
• Estimated lost sales: 8 units × $75 = $600
• Action: Increase size 16 allocation by 15-20%

Inventory Aging Analysis

Age Brackets:

• 0-30 days: Fresh inventory
• 31-60 days: Normal aging
• 61-90 days: Concerning (plan promotions)
• 91-120 days: Problem (markdown needed)
• 120+ days: Dead stock (liquidate)

Size-Specific Aging:

• If certain sizes consistently age faster, reduce allocation
• If sizes sell quickly, increase allocation

Building Size Distribution Models

Create data-driven allocation frameworks.

Historical Performance Model

Method:

• Analyze past 12 months of sales by size
• Calculate percentage of total sales per size
• Use percentages to allocate future orders

Example (Men's Dress Pants):

Historical Sales (Last 12 Months):

• Size 30: 45 units (5%)
• Size 32: 180 units (20%)
• Size 34: 225 units (25%)
• Size 36: 180 units (20%)
• Size 38: 135 units (15%)
• Size 40: 90 units (10%)
• Size 42: 45 units (5%)
• Total: 900 units

New Order (300 units):

• Size 30: 300 × 5% = 15 units
• Size 32: 300 × 20% = 60 units
• Size 34: 300 × 25% = 75 units
• Size 36: 300 × 20% = 60 units
• Size 38: 300 × 15% = 45 units
• Size 40: 300 × 10% = 30 units
• Size 42: 300 × 5% = 15 units

Adjusted Performance Model

Adjustments to Historical Model:

1. Stockout Adjustment:

• If size stocked out, actual demand was higher
• Increase allocation by estimated lost sales

Example:

• Size 34 sold 225 units but stocked out 3 times
• Estimated lost sales: 30 units
• Adjusted demand: 225 + 30 = 255 units (28% vs. 25%)
• New allocation: 300 × 28% = 84 units (vs. 75)

2. Markdown Adjustment:

• If size required markdowns, demand was lower
• Reduce allocation proportionally

Example:

• Size 42 sold 45 units, but 20 at 50% off
• True full-price demand: 25 units (3% vs. 5%)
• New allocation: 300 × 3% = 9 units (vs. 15)

3. Trend Adjustment:

• If size demand is growing/declining, adjust
• Compare recent 3 months to prior 9 months

Example:

• Size 32: Last 3 months 22% vs. prior 9 months 19%
• Upward trend: Increase allocation to 22%
• New allocation: 300 × 22% = 66 units (vs. 60)

Segmented Model

Different Allocations by Category:

Dress Shirts (Professional Market):

• 14.5-15: 10%
• 15.5-16: 40%
• 16.5-17: 35%
• 17.5-18: 15%

Casual Shirts (Broader Market):

• S: 15%
• M: 35%
• L: 30%
• XL: 15%
• XXL: 5%

Suits (Premium Market):

• 36-38: 15%
• 40-42: 50%
• 44-46: 25%
• 48+: 10%

Regional and Demographic Adjustments

Tailor allocations to specific markets.

Geographic Variations

Regional Body Type Differences:

Urban Markets (NYC, SF, LA):

• Trend toward smaller sizes
• Fitness-conscious demographics
• Slim fit preference
• Shift allocation 5-10% toward smaller sizes

Midwest/South:

• Larger average body size
• Regular/relaxed fit preference
• Shift allocation 5-10% toward larger sizes

Example Adjustment:

National Baseline (Dress Shirts):

• 15-15.5: 30%
• 16-16.5: 45%
• 17-17.5: 20%
• 18+: 5%

Urban Store Adjustment:

• 15-15.5: 35% (+5%)
• 16-16.5: 45% (same)
• 17-17.5: 17% (-3%)
• 18+: 3% (-2%)

Midwest Store Adjustment:

• 15-15.5: 25% (-5%)
• 16-16.5: 45% (same)
• 17-17.5: 23% (+3%)
• 18+: 7% (+2%)

Demographic Adjustments

Age Demographics:

Younger Market (18-35):

• Smaller sizes more common
• Slim fit preference
• Trend-conscious sizing

Mature Market (50+):

• Larger sizes more common
• Regular/relaxed fit preference
• Classic sizing

Implementing Predictive Analytics

Advanced techniques for sophisticated buyers.

Moving Average Method

Concept:

• Use average of recent periods to predict future
• Smooths out anomalies
• Responsive to trends

3-Month Moving Average Example:

Size 16 Shirt Sales:

• Month 1: 45 units
• Month 2: 52 units
• Month 3: 48 units
• 3-month average: (45 + 52 + 48) ÷ 3 = 48 units/month
• Forecast for next 3 months: 48 × 3 = 144 units

Seasonal Adjustment

Account for Seasonal Patterns:

Example (Suit Sales by Quarter):

• Q1 (Jan-Mar): 20% of annual sales
• Q2 (Apr-Jun): 30% of annual sales (wedding season)
• Q3 (Jul-Sep): 15% of annual sales
• Q4 (Oct-Dec): 35% of annual sales (holiday, back-to-work)

Application:

• If annual forecast is 1,000 suits
• Q2 order: 1,000 × 30% = 300 suits
• Apply size distribution to 300 units

Trend Analysis

Identify Growth/Decline Patterns:

Example (Size 34 Pants):

• 6 months ago: 18% of sales
• 3 months ago: 20% of sales
• Current: 22% of sales
• Trend: +2% per quarter
• Forecast next quarter: 24% of sales
• Adjust allocation accordingly

Tools and Systems

Technology enables data-driven buying.

Essential Tools

Point-of-Sale (POS) System:

• Captures sales by size automatically
• Real-time inventory tracking
• Sales reporting by size
• Essential foundation

Inventory Management Software:

• Tracks inventory levels by size
• Alerts for low stock
• Reorder point calculations
• Aging reports

Spreadsheet Analysis:

• Excel or Google Sheets
• Custom size analysis templates
• Flexible and accessible
• Good starting point

Advanced Analytics Platforms:

• Dedicated retail analytics software
• Predictive modeling
• Automated recommendations
• For larger operations

Building a Simple Analytics Dashboard

Key Metrics to Display:

• Sell-through % by size
• Inventory on hand by size
• Weeks of supply by size
• Stockout frequency by size
• Sales velocity by size
• Recommended reorder quantities

Update Frequency:

• Daily: Inventory levels, stockouts
• Weekly: Sell-through rates, sales velocity
• Monthly: Comprehensive analysis, reorder planning

Practical Implementation Guide

Step-by-step approach to data-driven size buying.

Phase 1: Data Collection (Month 1)

Actions:

• Set up size tracking in POS system
• Create spreadsheet templates
• Begin capturing sales by size
• Document current inventory by size
• Establish baseline metrics

Phase 2: Analysis (Months 2-3)

Actions:

• Analyze 2-3 months of data
• Calculate sell-through by size
• Identify stockouts and excess
• Create initial size distribution model
• Compare to industry benchmarks

Phase 3: Implementation (Month 4)

Actions:

• Apply data-driven model to next order
• Start conservatively (adjust 10-15% from gut feel)
• Document allocation decisions
• Set up tracking for results

Phase 4: Refinement (Months 5-6)

Actions:

• Measure results of data-driven order
• Compare to previous gut-feel orders
• Refine model based on results
• Increase confidence in data
• Expand to more categories

Phase 5: Optimization (Ongoing)

Actions:

• Continuously refine models
• Incorporate new data
• Adjust for market changes
• Automate where possible
• Share insights across team

Common Mistakes to Avoid

1. Insufficient Data:

• Making decisions on 1-2 months of data
• Seasonal anomalies skew results
• Solution: Minimum 3 months, ideally 12 months

2. Ignoring Stockouts:

• Only analyzing what sold, not what could have sold
• Underestimates true demand
• Solution: Track and adjust for stockouts

3. Over-Reliance on Averages:

• National averages don't fit local markets
• Your customer base is unique
• Solution: Use your own data, adjust for local factors

4. Analysis Paralysis:

• Waiting for perfect data before acting
• Missing ordering windows
• Solution: Start with available data, refine over time

5. Ignoring Qualitative Factors:

• Pure data without context
• Missing market shifts or trends
• Solution: Combine data with market knowledge

Measuring Success

Track improvement to validate approach.

Key Performance Indicators

Before Data-Driven Buying:

• Overall sell-through: 65%
• Markdown rate: 25% of inventory
• Stockout frequency: 15% of sizes
• Inventory turnover: 3x per year

After Data-Driven Buying (Target):

• Overall sell-through: 85%+ (20% improvement)
• Markdown rate: 10% of inventory (60% reduction)
• Stockout frequency: 5% of sizes (67% reduction)
• Inventory turnover: 4-5x per year (33-67% improvement)

Financial Impact:

• Reduced markdowns: +10-15% margin improvement
• Increased sales (fewer stockouts): +5-10% revenue
• Better inventory efficiency: -20-30% inventory investment
• Combined impact: 20-30% profitability improvement

Conclusion: From Gut Feel to Data-Driven Precision

The transition from intuitive to data-driven size buying represents one of the most impactful improvements a menswear retailer can make. By systematically collecting size data, analyzing performance, building allocation models, and continuously refining based on results, retailers can dramatically improve sell-through rates, reduce markdowns, minimize stockouts, and optimize inventory investment. The tools and techniques outlined in this guide provide a roadmap from basic data collection to sophisticated predictive analytics—scalable to any size operation and adaptable to any product category.

Key action steps:

• Start tracking today: Set up size-level sales and inventory tracking
• Analyze 3-12 months: Build baseline understanding of size performance
• Create distribution model: Use historical percentages as starting point
• Adjust for reality: Account for stockouts, markdowns, trends
• Segment by category: Different products have different size curves
• Localize allocation: Adjust for regional and demographic factors
• Implement gradually: Start with one category, expand as confidence grows
• Measure results: Track sell-through, markdowns, stockouts
• Refine continuously: Update models with new data
• Combine data and judgment: Analytics inform, experience guides

Remember that data-driven buying isn't about eliminating human judgment—it's about enhancing it with objective information. The most successful buyers combine analytical rigor with market knowledge, using data to validate instincts and identify blind spots. Start simple, build systematically, and let the results speak for themselves. Your inventory investment is too important to leave to guesswork.