Shopify Order Export CSV: Bundle Affinity Analysis

Introduction to Product Bundle Affinity Analysis

Understanding which products your customers buy together is one of the most powerful insights you can gain from your Shopify store data. Product bundle affinity analysis—also known as market basket analysis—reveals hidden patterns in purchasing behavior that can dramatically increase your average order value (AOV) and improve customer satisfaction.

When you know that customers who buy Product A are 80% likely to also purchase Product B, you can create strategic product bundles, optimize your product recommendation engine, and design more effective cross-sell campaigns. This isn't guesswork—it's data-driven decision making at its finest.

In this comprehensive tutorial, you'll learn exactly how to perform product bundle affinity analysis on your Shopify order data, interpret the results, and implement actionable strategies to boost revenue. Whether you're running a small boutique or managing a large e-commerce operation, these techniques will help you unlock value hiding in your transaction data.

Prerequisites and Data Requirements

What You'll Need Before Starting

To successfully complete this tutorial, ensure you have:

• Shopify Admin Access: You need permission to export order data from your Shopify store
• Sufficient Order History: At least 500-1000 completed orders for meaningful results (more is better)
• Multi-Product Orders: The analysis works best when customers regularly purchase multiple items per transaction
• Clean Product Data: Consistent product names and SKUs in your Shopify catalog
• MCP Analytics Account: Access to the Bundle Affinity Analysis service

Understanding Your Data Structure

Shopify order data typically includes these key fields for bundle analysis:

• Order ID: Unique identifier for each transaction
• Product Name/Title: The name of each product purchased
• Product SKU: Stock keeping unit (more reliable than names)
• Quantity: Number of units purchased
• Order Date: When the purchase was made
• Line Item Properties: Variants and customizations

The analysis algorithm groups products by Order ID to identify which items appear together in the same transaction, then calculates statistical measures of association strength.

Step 1: Export Your Shopify Order Data

The first step is extracting your order data from Shopify in a format suitable for analysis.

Exporting from Shopify Admin

1. Log into your Shopify Admin panel
2. Navigate to Orders from the left sidebar
3. Click the Export button in the top right corner
4. Select your export options:
   • Export: Choose "All orders" or a specific date range (last 6-12 months recommended)
   • Format: Select "CSV for Excel, Numbers, or other spreadsheet programs"
   • Export orders as: Choose "Plain CSV file"
5. Click Export orders
6. Wait for the download to complete (Shopify will email you for large exports)

Expected Output

You'll receive a CSV file named something like orders_export_1.csv. When opened, it should contain columns like:

Name,Email,Financial Status,Paid at,Fulfillment Status,Fulfilled at,Accepts Marketing,Currency,Subtotal,Shipping,Taxes,Total,Discount Code,Discount Amount,Shipping Method,Created at,Lineitem quantity,Lineitem name,Lineitem price,Lineitem compare at price,Lineitem sku,Lineitem requires shipping,Lineitem taxable,Lineitem fulfillment status
#1001,[email protected],paid,2024-01-15,fulfilled,2024-01-16,yes,USD,79.99,10.00,7.20,97.19,,,Standard,2024-01-15,1,Wireless Mouse,29.99,,MOUSE-001,TRUE,TRUE,fulfilled
#1001,[email protected],paid,2024-01-15,fulfilled,2024-01-16,yes,USD,79.99,10.00,7.20,97.19,,,Standard,2024-01-15,1,USB Keyboard,49.99,,KEYB-002,TRUE,TRUE,fulfilled

Notice how multi-product orders appear on multiple rows, with the same Order ID (#1001) repeated for each line item. This structure is perfect for bundle analysis.

Step 2: Prepare Your Data File

While Shopify exports are generally clean, a few preparation steps ensure optimal analysis results.

Data Cleaning Checklist

1. Remove Incomplete Orders

Filter out orders with status other than "paid" or "fulfilled" to focus on completed transactions:

• Remove pending orders
• Exclude refunded or cancelled orders (unless analyzing return patterns)
• Keep only orders with "Financial Status" = "paid"

2. Standardize Product Identifiers

Decide whether to use Product Names or SKUs (recommendation: use SKUs when available):

• SKUs are preferred: More consistent, less prone to spelling variations
• Product Names work: But ensure no duplicate names for different products
• Avoid mixing both in a single analysis

3. Handle Product Variants

Decide your granularity level:

• Option A: Treat variants separately (e.g., "T-Shirt Red" vs "T-Shirt Blue")
• Option B: Group variants together (e.g., all colors as "T-Shirt")

For most bundle analysis, Option B provides more actionable insights about product categories rather than specific variants.

4. Remove Test Orders

Filter out any test orders or internal purchases that might skew results.

Verification Step

After cleaning, verify your data contains:

• At least 2 essential columns: Order ID and Product Identifier
• Multiple line items for the same Order ID (multi-product purchases)
• No missing values in critical fields
• Consistent formatting throughout

Step 3: Upload Data to MCP Analytics

Now you're ready to analyze your prepared data using the Bundle Affinity Analysis tool.

Accessing the Analysis Tool

1. Navigate to the Bundle Affinity Analysis tool
2. Log in to your MCP Analytics account (or create one if needed)
3. You'll see the data upload interface

Uploading Your File

1. Click the "Choose File" or drag-and-drop area
2. Select your prepared Shopify orders CSV file
3. Wait for the upload progress bar to complete
4. The system will perform automatic validation checks

Column Mapping

The tool will attempt to auto-detect your columns, but verify the mapping:

• Transaction ID: Map to your "Name" or Order ID column
• Product/Item: Map to "Lineitem sku" or "Lineitem name"
• Quantity (optional): Map to "Lineitem quantity" if you want to weight by quantity
• Date (optional): Map to "Created at" for time-based filtering

Expected Confirmation

After successful upload and mapping, you should see a summary like:

✓ Data uploaded successfully
✓ 1,247 unique orders detected
✓ 89 unique products identified
✓ 3,421 total line items
✓ Average products per order: 2.74
Ready for analysis

Step 4: Configure Analysis Parameters

Bundle affinity analysis uses the Apriori algorithm to find association rules. Understanding the key parameters ensures you get meaningful, actionable results rather than noise.

Key Parameters Explained

1. Minimum Support

Definition: The minimum percentage of transactions that must contain an itemset for it to be considered.

Formula: Support(A) = (Transactions containing A) / (Total transactions)

Recommended Setting:

• Start with 1-2% for stores with 1000+ orders
• Use 5% for smaller datasets (under 500 orders)
• Too low = too many insignificant patterns
• Too high = miss interesting but less frequent bundles

2. Minimum Confidence

Definition: The minimum probability that product B is purchased when product A is purchased.

Formula: Confidence(A → B) = Support(A, B) / Support(A)

Recommended Setting:

• Start with 20-30% for exploratory analysis
• Use 50%+ for high-confidence bundle creation
• Lower values reveal weaker but still meaningful associations

3. Lift Threshold (Advanced)

Definition: How much more likely items are bought together compared to random chance.

Formula: Lift(A → B) = Confidence(A → B) / Support(B)

Interpretation:

• Lift = 1: No association (random)
• Lift > 1: Positive association (items bought together more than chance)
• Lift < 1: Negative association (items bought together less than chance)

Recommended Setting: Minimum lift of 1.2 to 1.5 filters for meaningful associations.

Configuration Example

For a Shopify store with 2,000 orders and 150 products, try these initial settings:

{
    "minimum_support": 0.015,     // 1.5% (30 orders minimum)
    "minimum_confidence": 0.25,   // 25% confidence
    "minimum_lift": 1.2,          // 20% above random chance
    "max_items_per_rule": 3       // Find bundles of up to 3 products
}

Click "Run Analysis" and the algorithm will process your data. This typically takes 30-120 seconds depending on dataset size.

Step 5: Interpret Association Rules

Once analysis completes, you'll receive a list of association rules ranked by various metrics. Understanding how to interpret these results is crucial for making strategic decisions.

Understanding the Results Table

Your results will appear in a table format like this:

Rule #  | Antecedent (If)      | Consequent (Then)    | Support | Confidence | Lift | Count
--------|----------------------|----------------------|---------|------------|------|------
1       | Yoga Mat             | Yoga Blocks          | 3.2%    | 45.3%      | 2.8  | 64
2       | Coffee Maker         | Coffee Filters       | 4.1%    | 67.2%      | 3.1  | 82
3       | Running Shoes        | Athletic Socks       | 2.8%    | 38.9%      | 2.2  | 56
4       | Laptop               | Laptop Case          | 5.3%    | 71.4%      | 2.9  | 106
5       | {Phone, Case}        | Screen Protector     | 1.9%    | 52.1%      | 3.5  | 38

Reading Each Rule

Rule #2 Example: Coffee Maker → Coffee Filters

• Support (4.1%): 4.1% of all orders contain both items (82 out of 2,000 orders)
• Confidence (67.2%): When customers buy a Coffee Maker, 67.2% also buy Coffee Filters
• Lift (3.1): Customers are 3.1x more likely to buy filters with a coffee maker than randomly
• Count (82): This pattern occurred in 82 transactions

Prioritizing Rules for Action

High-Value Bundle Opportunities

Look for rules with:

• High Confidence (>50%): Strong predictive power
• High Lift (>2.0): Meaningful association beyond chance
• Moderate Support (2-10%): Frequent enough to matter

These are your best candidates for creating fixed bundles or "Frequently Bought Together" recommendations.

Cross-Sell Opportunities

Rules with moderate confidence (25-50%) but high lift work well for:

• Product page recommendations
• Cart page upsells
• Post-purchase email campaigns

Complementary Product Discovery

Look for unexpected associations—products you wouldn't intuitively bundle but data shows customers buy together. These reveal customer use cases you might not have considered.

Statistical Significance

For deeper statistical validation of your findings, consider applying statistical significance testing to ensure patterns aren't due to random chance, especially with smaller datasets.

Step 6: Implement Bundle Strategies in Shopify

Analysis without action is wasted effort. Here's how to implement your findings in your Shopify store.

Strategy 1: Create Product Bundles

For High-Confidence Rules

Rules with 50%+ confidence and high lift are perfect for fixed bundles:

1. In Shopify Admin, go to Products → Add product
2. Create a new bundle product (e.g., "Coffee Starter Kit")
3. Set a bundled price (typically 10-15% discount)
4. Use a bundling app like "Bundle Builder" or "Bold Bundles" to link component products
5. Update product descriptions to highlight value and convenience

Strategy 2: Implement "Frequently Bought Together"

For moderate-confidence rules (25-50%), use dynamic recommendations:

1. Install a Shopify app like "Also Bought" or "LimeSpot"
2. Configure custom product recommendations based on your association rules
3. Display these on product pages and cart pages
4. A/B test recommendation placement for optimal conversion

Strategy 3: Optimize Product Pages

Update your product pages to naturally encourage bundle purchases:

• Add "Complete the Set" sections showing related items
• Include comparison tables showing bundle savings
• Use lifestyle images showing products used together
• Write copy highlighting complementary use cases

Strategy 4: Email Marketing Campaigns

Leverage bundle insights for targeted campaigns:

• Segment customers who bought Product A but not Product B
• Send personalized recommendations with bundle discount codes
• Create "You might also need" post-purchase sequences
• Time emails based on typical purchase cycles

Measuring Success

Track these metrics to validate your implementation:

• Average Order Value (AOV): Should increase 10-20% with effective bundling
• Attach Rate: Percentage of Product A purchases that include Product B
• Bundle Product Performance: Sales velocity of new bundle products
• Revenue Per Visitor: Overall improvement in conversion value

Re-run your bundle affinity analysis quarterly to identify evolving customer preferences and seasonal patterns.

Advanced Interpretation Techniques

Multi-Item Rules

Rules with multiple antecedents (e.g., {Laptop, Mouse} → Laptop Bag) reveal more complex buying patterns:

• These show what customers add after already having multiple items in cart
• Perfect for cart page recommendations
• Usually have lower support but can have very high confidence
• Indicate strong complementary product families

Negative Associations

Rules with lift < 1 show products bought together LESS than random chance:

• Might indicate competing products (different solutions to same problem)
• Useful for understanding product cannibalization
• Can inform inventory planning and product line strategy
• Help identify when you're offering too many similar options

Seasonal and Temporal Patterns

Run separate analyses for different time periods:

• Compare Q4 holiday shopping vs. rest of year
• Identify seasonal bundle opportunities
• Track how bundle patterns evolve over product lifecycle
• Adjust recommendations based on calendar events

Try Bundle Affinity Analysis Now

Ready to discover which products your customers are buying together? The MCP Analytics Bundle Affinity Analysis tool makes it easy to upload your Shopify data and get actionable insights in minutes.

No credit card required for your first analysis. Upload your CSV and see results immediately.

Next Steps and Further Learning

Expand Your Analytics Skills

Now that you understand bundle affinity analysis, explore these related techniques:

• AI-First Data Analysis Pipelines: Automate your Shopify analytics workflow with AI-powered insights
• AdaBoost for Predictive Analytics: Predict which customers are most likely to purchase specific bundles
• Customer Segmentation: Identify different customer segments with distinct bundle preferences
• Price Optimization: Determine optimal bundle pricing using elasticity analysis

Advanced Shopify Analytics

Take your Shopify data analysis further:

• Cohort analysis to track bundle adoption over time
• RFM segmentation for targeted bundle marketing
• Inventory forecasting based on bundle popularity
• Lifetime value modeling incorporating bundle purchases

Stay Updated

Bundle affinity patterns evolve with your product catalog and customer base:

• Re-run analysis monthly or quarterly
• Track metric changes after implementing bundle strategies
• Experiment with different parameter thresholds
• A/B test bundle presentations and pricing

Common Issues and Solutions

Issue 1: No Meaningful Rules Found

Symptoms: Analysis returns very few rules or none at all.

Possible Causes:

• Minimum support threshold set too high
• Most orders contain only single items
• Insufficient order history (under 500 orders)
• Too many unique products with low individual sales

Solutions:

• Lower minimum support to 0.5-1%
• Group product variants together to reduce unique item count
• Focus on best-selling product categories only
• Collect more data before re-analyzing
• Consider if your business model naturally supports bundling

Issue 2: Too Many Rules (Information Overload)

Symptoms: Hundreds or thousands of rules returned, difficult to prioritize.

Solutions:

• Increase minimum confidence to 40-50%
• Set minimum lift to 2.0 or higher
• Filter for rules with support > 2%
• Export results and sort by lift × confidence × support
• Focus on top 10-20 rules for initial implementation

Issue 3: Obvious or Trivial Associations

Symptoms: Rules like "Phone Case → Phone" that are too obvious to be useful.

Solutions:

• These are validation that the analysis works correctly
• Filter out rules where consequent is a required component of antecedent
• Look deeper in the results for non-obvious patterns
• Consider reversing obvious rules (Phone → Phone Case) for cross-sell opportunities

Issue 4: Data Upload Errors

Symptoms: File won't upload or validation fails.

Common Causes and Fixes:

• File too large: Split into multiple date ranges, analyze separately, combine insights
• Encoding issues: Save CSV as UTF-8 encoding
• Missing required columns: Ensure Order ID and Product fields are present
• Special characters: Remove or escape commas and quotes in product names

Issue 5: Inconsistent Results Over Time

Symptoms: Running analysis on same data produces different rules.

Explanation:

This usually means you're close to parameter thresholds—small rounding differences can include/exclude borderline rules. This is normal.

Solution:

• Focus on rules that consistently appear across analyses
• Use stricter thresholds for more stable results
• Rules that barely meet thresholds may not be robust anyway

Issue 6: Can't Implement Bundles in Shopify

Symptoms: Technical limitations preventing bundle creation.

Solutions:

• Shopify Lite/Basic: Use discount codes for manual bundles (BUY2SAVE15)
• Inventory tracking: Use bundling apps that sync component inventory
• Variant limits: Create separate bundle products instead of complex variants
• App conflicts: Test bundle apps in isolation to identify conflicts

Getting Help

If you encounter issues not covered here:

• Check the Bundle Affinity service documentation for detailed specifications
• Review your data export to ensure proper formatting
• Start with a small sample (100-200 orders) to test the process
• Contact MCP Analytics support with specific error messages

Conclusion

Product bundle affinity analysis transforms your Shopify order data from a simple transaction log into a strategic asset. By understanding which products customers naturally buy together, you can create compelling bundles, optimize cross-sell opportunities, and increase average order value.

The six-step process outlined in this tutorial—exporting data, preparing files, uploading to MCP Analytics, configuring parameters, interpreting results, and implementing strategies—provides a repeatable framework for ongoing optimization.

Remember that bundle analysis isn't a one-time exercise. Customer preferences evolve, new products launch, and seasonal patterns emerge. Make this analysis part of your quarterly analytics routine to stay ahead of changing purchase behaviors.

Start with your most straightforward, high-confidence rules for quick wins, then progressively expand to more sophisticated bundling strategies. Track your results, iterate based on performance data, and watch your average order value grow.

Ready to uncover the hidden bundle opportunities in your Shopify store? Run your first bundle affinity analysis now and start making data-driven decisions about your product offerings today.