Analyzing Bundle Purchase Patterns

🧩 𝐌𝐮𝐥𝐭𝐢-𝐏𝐫𝐨𝐝𝐮𝐜𝐭 𝐏𝐮𝐫𝐜𝐡𝐚𝐬𝐞 𝐏𝐚𝐭𝐡 𝐀𝐧𝐚𝐥𝐲𝐬𝐢𝐬 𝐔𝐬𝐢𝐧𝐠 𝐆𝐀𝟒 + 𝐁𝐢𝐠𝐐𝐮𝐞𝐫𝐲 In e-commerce, not all purchase paths are created equal. Some products consistently lead to higher AOV (Average Order Value) or LTV (Customer Lifetime Value), but standard reports don’t tell you which. With GA4 + BigQuery, you can uncover those high-performing product paths and optimize your merchandising and conversion strategy accordingly. 🔹 𝗦𝘁𝗲𝗽 𝟭: 𝗠𝗮𝗽 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗣𝗮𝘁𝗵𝘀 𝗳𝗿𝗼𝗺 𝗔𝗱𝗱-𝘁𝗼-𝗖𝗮𝗿𝘁 → 𝗣𝘂𝗿𝗰𝗵𝗮𝘀𝗲 Use add_to_cart and purchase events from GA4’s BigQuery export to reconstruct multi-product sequences: 𝐒̲𝐐̲𝐋̲ ̲𝐐̲𝐮̲𝐞̲𝐫̲𝐲̲ ̲𝐄̲𝐱̲𝐚̲𝐦̲𝐩̲𝐥̲𝐞̲:̲ 𝘚𝘌𝘓𝘌𝘊𝘛  𝘶𝘴𝘦𝘳_𝘱𝘴𝘦𝘶𝘥𝘰_𝘪𝘥,  𝘈𝘙𝘙𝘈𝘠_𝘈𝘎𝘎(𝘱𝘳𝘰𝘥𝘶𝘤𝘵_𝘪𝘥 𝘖𝘙𝘋𝘌𝘙 𝘉𝘠 𝘦𝘷𝘦𝘯𝘵_𝘵𝘪𝘮𝘦𝘴𝘵𝘢𝘮𝘱) 𝘈𝘚 𝘱𝘳𝘰𝘥𝘶𝘤𝘵_𝘱𝘢𝘵𝘩 𝘍𝘙𝘖𝘔  `𝘱𝘳𝘰𝘫𝘦𝘤𝘵.𝘥𝘢𝘵𝘢𝘴𝘦𝘵.𝘦𝘷𝘦𝘯𝘵𝘴_*`,  𝘜𝘕𝘕𝘌𝘚𝘛(𝘪𝘵𝘦𝘮𝘴) 𝘈𝘚 𝘱𝘳𝘰𝘥𝘶𝘤𝘵 𝘞𝘏𝘌𝘙𝘌  𝘦𝘷𝘦𝘯𝘵_𝘯𝘢𝘮𝘦 𝘐𝘕 ('𝘢𝘥𝘥_𝘵𝘰_𝘤𝘢𝘳𝘵', '𝘱𝘶𝘳𝘤𝘩𝘢𝘴𝘦') 𝘎𝘙𝘖𝘜𝘗 𝘉𝘠  𝘶𝘴𝘦𝘳_𝘱𝘴𝘦𝘶𝘥𝘰_𝘪𝘥 This shows you which product combos users add and buy in the same session or over time. 🔹 𝗦𝘁𝗲𝗽 𝟮: 𝗜𝗱𝗲𝗻𝘁𝗶𝗳𝘆 𝗛𝗶𝗴𝗵-𝗩𝗮𝗹𝘂𝗲 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗖𝗼𝗺𝗯𝗶𝗻𝗮𝘁𝗶𝗼𝗻𝘀 𝐀̲𝐧̲𝐚̲𝐥̲𝐲̲𝐳̲𝐞̲:̲ • Which product sequences most often result in purchases? • Which combinations lead to above-average AOV or repeat purchases? • Are there any gateway products (frequently purchased first)? 𝐔̲𝐬̲𝐞̲ ̲𝐦̲𝐞̲𝐭̲𝐫̲𝐢̲𝐜̲𝐬̲ ̲𝐥̲𝐢̲𝐤̲𝐞̲:̲ • AOV per product bundle • LTV of users who start with specific items • Frequency of “bought together” relationships 🔹 𝗦𝘁𝗲𝗽 𝟯: 𝗙𝗲𝗲𝗱 𝗜𝗻𝘀𝗶𝗴𝗵𝘁𝘀 𝗶𝗻𝘁𝗼 𝗖𝗥𝗢 𝗼𝗿 𝗥𝗲𝗰𝗼𝗺𝗺𝗲𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗘𝗻𝗴𝗶𝗻𝗲𝘀 Once you find your best-performing bundles: • Use the data to optimize product placements and cross-sells • Test product sequencing on PDPs and category pages • Feed insights into recommendation models (e.g., “Users who added X also buy Y”) 🚀 Real-world Outcomes: • Improved bundle targeting = higher cart sizes • Smarter personalization strategies = more returning customers • Data-driven product discovery = better merchandising decisions 📊 When you move beyond basic ecommerce reports and get granular with product path analysis, you turn insights into serious revenue. #EcommerceAnalytics #GA4 #BigQuery #DigitalAnalytics #MarTech #ProductAnalytics #CustomerJourney #RecommendationEngine #MerchandisingStrategy #DataDrivenMarketing #SQLforMarketing

Amazon's "Frequently Bought Together" generated $35 billion last year. But the algorithm isn't what you think. Everyone assumes it shows what other people bought. Wrong. It shows what will make the current cart profitable. Here's how it actually works: Step 1: Calculate margin on current item Step 2: Find items that increase total basket margin Step 3: Filter for return rate impact Step 4: Show combination most likely to never come back Example: You're buying a $30 phone case (40% margin). It COULD show: Screen protector ($8, lots of people buy it) It ACTUALLY shows: Wireless charger ($45, high margin, never returned) The really dirty secret? They'll sometimes show combinations they know work poorly together. Why? Because when you buy incompatible items, you keep both and buy a third item to fix it. Customer buys laptop + wrong charger = Customer keeps both, buys right charger = 3 sales instead of 2. But here's the masterclass part: They track "bundle regret rate" - how often people return one item from a bundle. If you return just the add-on, they still won. They got the original sale plus free data about what doesn't work. So what's the lesson? For your store: - Stop showing "works well with". Start showing "solves the next problem". - Track bundle return patterns. What gets kept vs. returned? - Price bundles so returning one item feels like losing value - Create intentional friction: "These items were selected together for a reason" The best upsell is what makes returning your product feel incomplete.

𝗔𝗺𝗮𝘇𝗼𝗻 𝗸𝗻𝗼𝘄𝘀 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂’𝗿𝗲 𝗴𝗼𝗶𝗻𝗴 𝘁𝗼 𝗯𝘂𝘆 𝗯𝗲𝗳𝗼𝗿𝗲 𝘆𝗼𝘂 𝗱𝗼. That’s not magic. That’s Market Basket Analysis. And it’s one of the most quietly powerful techniques in data analytics. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗶𝘁? Market Basket Analysis looks at what products customers buy together — and uses those patterns to drive smarter business decisions. Three metrics power the whole thing: → Support — how often do these two products appear together? → Confidence — if someone buys A, how likely are they to buy B? → Lift — is this pairing actually meaningful, or just a coincidence? 𝗪𝗵𝗮𝘁 𝗿𝗲𝗮𝗹 𝗽𝗿𝗼𝗯𝗹𝗲𝗺𝘀 𝗱𝗼𝗲𝘀 𝗶𝘁 𝘀𝗼𝗹𝘃𝗲? 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗿𝗲𝗰𝗼𝗺𝗺𝗲𝗻𝗱𝗮𝘁𝗶𝗼𝗻𝘀 “Customers who bought this also bought…” isn’t a guess. It’s a lift score above 1.0 telling you the pairing is statistically meaningful. 𝗦𝘁𝗼𝗿𝗲 𝗮𝗻𝗱 𝘄𝗲𝗯𝘀𝗶𝘁𝗲 𝗹𝗮𝘆𝗼𝘂𝘁 If bread and butter are always bought together, they shouldn’t be on opposite ends of the store. Same logic applies to your website’s product placement. 𝗕𝘂𝗻𝗱𝗹𝗶𝗻𝗴 𝗮𝗻𝗱 𝗽𝗿𝗼𝗺𝗼𝘁𝗶𝗼𝗻𝘀 Instead of blanket discounts, you offer targeted bundles based on what customers already buy together. Higher conversion. Lower cost. 𝗜𝗻𝘃𝗲𝗻𝘁𝗼𝗿𝘆 𝗽𝗹𝗮𝗻𝗻𝗶𝗻𝗴 If two products are frequently bought together, running out of one kills the sale of both. Procurement teams need to know this. 𝗖𝗿𝗼𝘀𝘀-𝘀𝗲𝗹𝗹𝗶𝗻𝗴 Sales teams stop guessing what to pitch next. The data tells them which product combinations have the highest confidence. 𝗜 𝗯𝘂𝗶𝗹𝘁 𝘁𝗵𝗶𝘀 𝗳𝗼𝗿 𝘁𝗵𝗲 𝗦𝗵𝗼𝗽𝗦𝗽𝗵𝗲𝗿𝗲 𝗽𝗿𝗼𝗷𝗲𝗰𝘁. Analysing website orders across product pairs — calculating support, confidence, and lift — then going a layer deeper into cross-category pairings and average basket value. The SQL alone tells you which products belong together, which categories cross-sell naturally, and how much each basket is worth on average. That’s not just analysis. That’s a roadmap for revenue. Every e-commerce business is sitting on this data. Most aren’t using it. Full project: github.com/Nte-Daniels #DataAnalytics #SQL #DataEngineering #Datafam

Take a guess: Which two products cross-sell best with Coke at the major US pharmacy chains? Keep reading and learn how I did the analysis. I worked on trade spend optimization at Coca-Cola with a simple goal: generate more revenue from the same shelf space. Here's exactly how I did it. The most impactful analysis I worked on at Coca-Cola had nothing to do with dashboards. The solution? Market basket analysis. We wanted to understand what people were buying with Coca-Cola. In real stores, with real purchasing behavior. One result still sticks with me. In two major pharmacy chains in the US, two of the top products purchased alongside Coca-Cola were Jim Beam and Jack Daniels. Was that your guess? Not likely. Most people say chips, chocolate, any of your other favorite junk foods. I found this really ironic, given these are pharmacy chains, not liquor stores. That insight mattered. It told us something simple but powerful. If people are buying these products together, placement matters. So we tested a small change. We placed Coca-Cola closer to those products in-store. That one decision unlocked tens of millions of dollars in revenue. Not because we sold more products overall, but because we made it easier for customers to buy what they already wanted together. That’s the real power of market basket analysis. The technique itself is relatively straightforward. The impact comes from knowing what the results mean, how to interpret them, and how to act on them. This week in an advanced Next-Level Tableau class, we recreated that kind of analysis using grocery store purchase data. The GIF shows how quickly these patterns jump out when you visualize them properly. Once you see relationships instead of rows, you stop asking "What sold?" and start asking "What should be next to what?" That’s where analysis starts to change outcomes. I’d love to hear, what kinds of use cases could you apply market basket analysis to?

DTC brands live and die by LTV, and yet 90% of brands are relying on near-identical email followups with diminishing results. Do this instead (tested across $350M of attributable revenue): 1/ Find your hero product Start by analyzing order volume across SKUs. You’re looking for the SKU that consistently shows up across the majority of purchases. Ask questions like: - What % of total orders include this product? - Are customers buying it solo or with something else? - Which 2-product combos appear most frequently? In one recent example, 90% of orders were 2-product combos and nearly all included the same hero SKU. 2/ Choose a low-COGS gift that complements the hero Look at what products are most frequently paired with your hero SKU. - Is there a natural second product that adds to the user experience? - Does it solve a related problem or fit into the same routine? - Can it be added to cart without adding operational complexity? Ideally, you want something that: - Costs <$3 in COGS - Introduces a new product line or increases usage frequency - Is small/light enough to ship easily 3/ Bundle the two into a clean, high-perceived value offer Once you know the hero and the gift, create an offer that pairs the two clearly. For example: “Get [Hero Product] + a Free [Complimentary Gift] — limited-time only.” The job of this offer is to improve repeat purchase odds by introducing more SKUs on Day 1. 4/ Fit the offer into your CAC math The biggest mistake brands make here is forgetting unit economics. Even if the gift only adds $2–3 in COGS, you need to ensure that your cost caps still hold and you're maintaining efficiency at your target ROAS. If Meta can’t hit your CPA with this new offer, revisit price or positioning. But in our experience, offers like these tend to improve performance, especially if the added product has strong perceived value. If you want me to analyze your order data, identify your hero SKU pairing, and build a high-LTV offer like this, shoot me a DM or book a call with me.

We're testing something that makes most DTC founders nervous. For a pet wellness brand doing 8-figures annually, we're moving their popular 3-pack bundle from last position to middle position. And moving the 2-pack from middle to third. The context: This brand sells calming diffusers for anxious pets. Their current setup frames packages as room coverage: 1 room, 2 rooms, 3 rooms. The hypothesis? Your top option gets the most views, but the second position still outperforms the third in visibility. So why waste that prime real estate on your lowest-margin offer? Bundle psychology is something we've spent years studying. Putting the 3x bundle second helps in two ways: 1️⃣ It boosts AOV - more people end up choosing that higher-value option. 2️⃣ It can improve CVR - it anchors the 1-pack price and makes it feel like a better deal. Most customers scan left to right, but their eyes focus on that second position early in their decision process. If your highest-value bundle is buried in position 3, you're leaving money behind. It’s simple. The test setup: ∙ Original: Starter Pack (1 room), Best Seller (2 rooms), Biggest Savings (3 rooms) • Variant: Starter Pack (1 room), Biggest Savings (3 rooms), Best Seller (2 rooms) We're tracking conversion rate by package and overall AOV impact. Early hypothesis: Higher AOV with minimal conversion rate impact. The counterargument? Some customers might feel "pushed" into bigger purchases and drop off entirely. Especially for pet parents who might only need coverage for just one or two rooms.. That's exactly why we test instead of guess. Your bundle order isn't random. It's psychology.

I analysed the top 50 Shopify supplement brands. 💊 Why? They’re the hardest working marketers. 70% use the same strategy to boost revenue: Product bundles. I've used this simple tactic before and it doubled ROAS for paid ads. Here's how to build winning bundles using Shopify Sidekick, their AI chat tool: 1. Find Natural Pairs Use Shopify Sidekick to check your "frequently bought together" data Look for products bought together 30% of the time or more Pick items that work well together, not competing products 2. Spot Multi-Buy Patterns Ask Sidekick which products customers buy in bulk Check your average quantity per order Focus on items where people buy 2 or more units 3. Build Smart Bundles Price bundles 10-15% less than buying items on their own Create dedicated pages for each bundle Run ads to bundle pages Target 30% higher order values Quick Start: Test this with the free Shopify Bundles app. It creates virtual bundles your team can fulfil. Bottom Line: Brands using this strategy see 30% higher order values and double the return on ad spend. Most see results within 30 days of launch. The best part? Your best customers already show you which bundles to create!

🔥 Still pouring ad spend into single-product sales? You're burning margin. And it's killing your scale. Strong statement? Yes. Backed by data? 100%. After analyzing 100+ eCom brands spending $50K+/month, the results are crystal clear: 🚫 Single-Product Economics > Higher CAC relative to AOV > Fixed shipping eats more margin > Lower purchase frequency > Higher return rates > Fewer creative angles = faster fatigue ✅ Bundle Economics > Higher AOV absorbs rising CACs > Shipping cost spread across SKUs > More value = fewer returns > Easier to position as a full solution > Way more ad angles to test Here’s the 5-step playbook that keeps winning: 1. Analyze Purchase Behavior > Find natural bundles from past orders • Split by first-time vs repeat buyers • Set your bundle value target (e.g. $90–120) 2. Build a BYOB (Build Your Own Bundle) Page > Tiered pricing (e.g. 3+ = 10% off) • Add logic: cross-sell by use case or category • Make it fun to build - not a chore 3. Optimize the Cart Experience > Add milestone progress (free gift at $X) • Trigger bundle suggestions when single items are added • Use scarcity-driven bundles (e.g. “Only 250 left”) 4. Align Your Ad Creative > Stop selling features - sell the bundle value • Lead with savings, convenience, and transformation • Position bundles as “the complete solution” 5. Create VIP-Only Bundles > Offer pre-applied discounts for high-value customers • Build email/SMS flows that trigger bundle offers post-purchase • Reward loyalty with early access and exclusives The bottom line: If your growth has plateaued and your AOV isn’t climbing, bundling isn’t just a nice-to-have... it’s your next best lever. It improves margins, strengthens retention, and gives your ad account a lot more room to breathe. ---- Spending $50K+/month? Want a second set of eyes on how your brand’s performing? Drop me a “AUDIT” and I’ll personally review: → Your ad creative and messaging → Media buying infrastructure → Landing pages and checkout flow

🚀 How can a Retail/CPG company can boost revenues using Sequential Market Basket Analysis 🛒 Did you know that understanding what customers buy next is just as important as knowing what they buy together? 🔍 Let’s take a look at how Company X, a leader in Retail/CPG, transformed their bottom line using Sequential Market Basket Analysis : 🎯 The Challenge : Company X noticed that while their traditional market basket analysis identified frequently purchased item pairs (e.g., bread & butter), they were missing insights about the order in which purchases happened. For example, customers were buying laptops, but accessories like mouse pads weren’t selling until much later—or not at all. 💡 The Solution : By adopting Sequential Market Basket Analysis, they started analyzing the temporal patterns of purchases: - Customers bought laptops on Day 1. - Many returned to buy laptop bags on Day 7. - Some returned for mouse pads on Day 15. With this insight, they : 1️⃣ Introduced timely recommendations on their e-commerce platform (e.g., “Customers who purchased a laptop often buy a bag within a week!”). 2️⃣ Sent personalized emails nudging customers to buy complementary items. 3️⃣ Optimized their promotion strategy by timing discounts to align with expected purchase windows. 📈 The Results : 1️⃣ 25% revenue increase for high-value accessory items. 2️⃣Improved customer satisfaction as buyers found relevant suggestions when they needed them. 3️⃣Inventory efficiency gains, reducing overstock of unsold items. 🔥 Why It Works: Unlike traditional market basket analysis, sequential analysis identifies purchase pathways, helping businesses predict and influence the next logical step in the customer journey. 👉 If you're in Retail/CPG Analytics, ask yourself : ✅ Are you leveraging purchase sequences to boost sales? ✅ Are your recommendations based on when customers buy, not just what they buy together? 💡 Pro Tip: Tools like sequence mining algorithms (PrefixSpan, AprioriAll) or Markov Chains can help unlock these insights. Temporal data analysis is a different ball game all together! #sequentialpatterns #marketbasketanalysis #retailCPG #experientiallearning #datascience #apriori