This guide covers everything DTC brands need to know about ecommerce product recommendations and ecommerce filters on Shopify and Shopify Plus: the four recommendation types that work, when to use Shopify apps vs custom recommendation engines, how to build faceted filtering that actually converts, and the AOV math that makes both pay for themselves within 6 months. By the end, you'll have a clear implementation framework, whether you DIY with apps or build custom engines as part of internal tools and automation.
What Are Ecommerce Product Recommendations?
Ecommerce product recommendations are personalized product suggestions shown to shoppers throughout their journey, on product pages, in carts, at checkout, and post-purchase. Recommendations work because they reduce decision fatigue: instead of forcing shoppers to navigate hundreds of products themselves, you surface the most relevant ones based on what they're already looking at or buying.
The four most effective ecommerce product recommendation patterns:
1. Frequently Bought Together
Shows products commonly purchased alongside the current product. Typically appears on product pages just below the add-to-cart button. Frequently Bought Together is the highest-ROI recommendation pattern for most DTC stores, it lifts AOV by 8–15% when implemented correctly.
2. Customers Also Viewed / Customers Also Bought
Cross-sells based on browsing or purchase patterns of similar shoppers. Strong for top-of-funnel and middle-of-funnel discovery. Lifts cart-add rate by 5–10%.
3. Recommended For You / You May Like
Personalized recommendations based on individual browsing history, past purchases, or behavioral signals. Most powerful when shopper has 5+ sessions of history. Lifts AOV by 5–15%.
4. Complete the Look / Bundle Offers
Style-based or category-based product bundling, especially common in apparel, beauty, and home goods. Combines multiple SKUs into a discounted bundle. Lifts AOV by 10–25%.
Stores using all four recommendation types together typically see 20–30% AOV lift without changing ad spend.
At Conversion Design, we implement these recommendation patterns across Shopify and Shopify Plus stores, either through optimized app stacks for brands at <$1M/year revenue, or via custom recommendation engines for brands at $1M+/year where apps become a PageSpeed and cost liability.
What Are Ecommerce Filters?
Ecommerce filters help shoppers narrow down product catalogs to find what they want quickly. Without filters, shoppers facing 50+ products on a collection page often leave instead of scrolling endlessly. Strong filtering reduces bounce rate by 15–30% and lifts conversion rate by 8–20% on collection pages.
The 6 Filter Types Every Shopify Store Needs
| Filter Type | Use Case | Conversion Impact |
|---|---|---|
| Price range | Universal, every category | +10–15% conversion |
| Availability | Show in-stock first | +8–12% conversion |
| Rating | "4 stars and up" | +6–10% conversion |
| Bestsellers | Sort or filter by popularity | +5–8% conversion |
| Category-specific | Size, color, material, etc. | +15–25% conversion |
| Custom attributes | Brand-specific (vegan, certified, etc.) | +10–18% conversion |
Faceted Filtering vs Basic Filtering
Basic filtering lets shoppers select one filter at a time. Faceted filtering lets shoppers combine multiple filters simultaneously, e.g., "red AND size medium AND under €50 AND in stock."
The difference shows up in revenue. Best-in-class faceted filtering reduces bounce rate by 25–40% on collection pages with 50+ products vs basic filtering.
Default Shopify filtering is basic. Custom faceted filtering on Shopify Plus is one of the highest-ROI internal tools investments for DTC brands at $1M+/year revenue.
The AOV Math, Why Recommendations Pay for Themselves
The revenue case for ecommerce product recommendations is mathematically simple:
Example: $500K/year DTC store
- Current AOV: €80
- Monthly orders: 520
- Monthly revenue: €41,600
After implementing recommendations (15% AOV lift)
- New AOV: €92
- Monthly orders: 520 (unchanged)
- Monthly revenue: €47,840
- Monthly lift: €6,240
- Annual lift: €74,880
Investment cost
- Shopify recommendation app: €100/month = €1,200/year
- OR custom recommendation engine: €10,000 one-time
ROI
- Apps: €74,880 lift / €1,200 cost = 62x ROI
- Custom engine: Pays back in 1.6 months, generates €60K+/year ongoing
For most DTC brands at $500K–$5M/year revenue, ecommerce product recommendations are the highest-ROI single change you can make to your Shopify store.
Where to Place Product Recommendations on Shopify
The placement of recommendations matters as much as the algorithm. Here's the proven hierarchy:
🛍️ Product Detail Page (Highest Priority)
- Above-the-fold: Don't crowd. Keep the focus on the product.
- Just below add-to-cart: "Frequently Bought Together", the highest-ROI placement.
- Mid-page: "Complete the Look" or "Customers Also Viewed."
- Bottom of page: "Recommended For You" with broader category recommendations.
🛒 Cart / Cart Drawer (Critical for AOV)
- In-cart upsells: "Add this for free shipping" or "Customers also added..."
- Bundle offers: "Save 10% when you add..."
- Recommended add-ons: Lower-priced, high-margin items.
✅ Checkout Page (Shopify Plus Only)
Shopify Plus's Checkout Extensibility allows custom recommendation slots in checkout. Most powerful placement because shopper has already committed to buying.
Checkout upsells lift AOV by 8–18% on Shopify Plus stores.
🎉 Post-Purchase / Order Confirmation
Post-purchase upsells are the highest-converting recommendation placement, shopper has just bought, payment is captured, friction is zero. One-click add to existing order.
Post-purchase upsells lift AOV by 12–25% with no impact on initial conversion rate.
📧 Email & SMS (Outside Shopify)
- Cart abandonment emails: Show abandoned products + complementary recommendations.
- Post-purchase emails: Recommend products based on what they bought.
- Browse abandonment: Trigger emails based on product viewing behavior.
The combined revenue impact of recommendations across product page + cart + checkout + post-purchase + email typically lifts total store revenue by 25–40% within 90 days.
Shopify Recommendation Apps vs Custom Engines
The biggest decision in implementing ecommerce product recommendations on Shopify: app or custom build.
Top Shopify Recommendation Apps
| App | Pricing | Best For |
|---|---|---|
| Shopify Search & Discovery | Free | Basic recommendations, brands <$500K/year |
| LimeSpot Personalizer | €18–€480/month | Mid-market DTC, AI-driven recommendations |
| Rebuy | €99–€999/month | Mid-to-large DTC, smart cart upsells |
| Searchspring | €600–€2,500/month | Enterprise DTC, advanced search + filtering |
| Boost AI Search | €29–€399/month | Mid-market, search-focused brands |
| AfterSell | €34–€199/month | Post-purchase upsell specialist |
| ReConvert | €0–€199/month | Post-purchase upsell, freemium model |
Custom Recommendation Engine
Custom Shopify recommendation engines built on Shopify Plus APIs typically cost €8,000–€25,000 one-time, with ongoing maintenance from €500/month.
App vs Custom, Decision Framework
| Factor | Use App | Build Custom |
|---|---|---|
| Revenue | <$1M/year | $1M+/year |
| PageSpeed sensitivity | Low | High |
| Algorithm control needs | Generic OK | Need brand-specific logic |
| Per-impression costs | Acceptable | Want to eliminate |
| App stack already large | Adding 1 more is fine | Already too many apps |
| Tech team available | No | Yes (or hire agency) |
Most DTC brands at <$1M/year do well with apps. At $1M+/year, custom recommendation engines pay back within 6–12 months through PageSpeed improvements, eliminated per-impression costs, and brand-specific algorithm performance.
How to Build Custom Faceted Filtering on Shopify
For brands at $1M+/year revenue with 100+ SKUs, default Shopify filtering becomes a conversion bottleneck. Custom faceted filtering is the solution.
Why Default Shopify Filtering Falls Short
- Limited filter types (often missing custom attributes)
- Slow performance on large catalogs
- Poor mobile UX (filters often hidden in modals)
- Limited multi-filter combination logic
- No "available combinations" highlighting (shoppers select filters that return zero results)
What Custom Faceted Filtering Adds
- Unlimited filter dimensions including custom metafields
- Real-time count updates ("23 products" → updates as you filter)
- Available combinations highlighting (gray out filter options that return zero)
- Mobile-first filter UX with sticky filter bars
- URL-based filter state for shareability and SEO
- Performance, sub-200ms filter responses even on 1,000+ SKU catalogs
Implementation Approaches
| Approach | Cost | Timeline | Best For |
|---|---|---|---|
| Shopify Search & Discovery + custom UI | €3,000–€8,000 | 2–4 weeks | Mid-market with <500 SKUs |
| Searchspring or Algolia integration | €2,000–€5,000 build + €600–€2,500/mo | 4–6 weeks | Enterprise with 500+ SKUs |
| Fully custom faceted filtering | €8,000–€25,000 | 6–10 weeks | Premium brands wanting full control |
Conversion Design builds custom faceted filtering as part of internal tools and automation services from €3,000 (filtering only) or as part of full Shopify Plus builds from €5,600.
Common Ecommerce Product Recommendation Mistakes
After auditing 200+ DTC brands, these are the recommendation mistakes that kill AOV lift:
❌ Mistake 1: Showing 8+ Recommendations
More recommendations = lower conversion per recommendation. The sweet spot is 3–6 recommendations per slot. Beyond that, decision fatigue takes over.
❌ Mistake 2: Same Recommendations Everywhere
Showing the same "Frequently Bought Together" on every product is lazy implementation. Real recommendations are product-specific based on actual purchase data.
❌ Mistake 3: Recommendations Below the Fold on Mobile
70%+ of DTC traffic is mobile. Recommendations buried 3 scrolls below the add-to-cart get <5% engagement on mobile vs 25–40% above-the-fold.
❌ Mistake 4: Slow-Loading Recommendation Apps
Many recommendation apps add 800–1,500ms to page load. The conversion lift from recommendations gets erased by the PageSpeed loss. Test before/after PageSpeed when adding any app.
❌ Mistake 5: No Post-Purchase Upsells
Post-purchase is the highest-converting recommendation placement. Skipping it leaves 12–25% AOV lift on the table. Use Shopify Plus Checkout Extensibility or apps like ReConvert/AfterSell.
❌ Mistake 6: Generic "You May Also Like"
Vague recommendations underperform specific ones. "Customers who bought X also bought Y" converts 2–3x better than "You May Also Like" because specificity builds trust.
❌ Mistake 7: Ignoring Filter UX on Mobile
Mobile filtering is where most stores leak revenue. Filters hidden behind a "Filter" button (with no count visible) get 60% lower engagement than sticky filter chips visible on scroll.
Common Ecommerce Filter Mistakes
The 7 filter mistakes we see most often in Shopify store audits:
❌ Mistake 1: Too Many Filters
8+ filters paralyze shoppers. 3–6 filters maximum per collection page, with the most-used ones at the top.
❌ Mistake 2: Not Showing Result Counts
"Color: Red" without "(23 products)" forces shoppers to click and discover. Showing real-time counts dramatically improves filter usage.
❌ Mistake 3: Allowing Zero-Result Combinations
If a shopper can select "size XL + color teal + under €30" and get zero results, you've wasted a click. Smart filtering grays out options that would return zero.
❌ Mistake 4: Filters That Don't Persist on Pagination
Shopper filters by "size medium," scrolls to page 2, filter resets. Major UX failure. Filter state must persist via URL parameters.
❌ Mistake 5: Mobile Filters Hidden Behind a Button
On mobile, filters should be sticky filter chips at the top of collection pages, not buried behind a modal "Filter" button.
❌ Mistake 6: No Sorting Options
Filtering narrows. Sorting orders. Both matter. Common sorts: bestseller, price low-high, price high-low, newest, rating.
❌ Mistake 7: Slow Filter Response Times
Filter updates that take 2+ seconds feel broken. Custom faceted filtering on Shopify Plus delivers sub-200ms responses even on 1,000+ SKU catalogs.
Industry-Specific Recommendation & Filter Strategies
Different DTC categories need different approaches:
💊 Supplements & Nutrition
Top filters: Goal (sleep, energy, recovery), format (capsule/powder/liquid), dietary (vegan, gluten-free), serving size, price.
Top recommendations: Frequently Bought Together (stack supplements), Bundle Offers (subscriptions), Post-Purchase Upsells (related goals).
👕 Apparel & Fashion
Top filters: Size, color, fit, material, occasion, price.
Top recommendations: Complete the Look, Customers Also Viewed, Size-Based Bundle Offers.
💄 Beauty & Cosmetics
Top filters: Skin type, concern, ingredient (paraben-free, vegan), shade, price.
Top recommendations: Frequently Bought Together (full routine), Customers Also Viewed, Personalized "For Your Skin Type."
🏋️ Fitness & Wellness
Top filters: Goal, level (beginner/advanced), equipment needed, format, price.
Top recommendations: Bundle Offers (program packages), Frequently Bought Together, Post-Purchase Upsells.
🍔 Food & Beverage
Top filters: Category, dietary, flavor, size, price.
Top recommendations: Frequently Bought Together (meal pairings), Subscription Bundles, Post-Purchase Upsells.
🐶 Pet Products
Top filters: Pet type, size, age, food type, dietary needs, price.
Top recommendations: Frequently Bought Together (food + accessories), Bundle Offers, Subscription Recommendations.
🏠 Home Goods
Top filters: Room, color, material, size, price.
Top recommendations: Complete the Look (rooms), Customers Also Viewed, Bundle Offers.
📱 Electronics & Gadgets
Top filters: Brand, compatibility, feature set, price.
Top recommendations: Frequently Bought Together (accessories), Compatible Products, Bundle Offers.
In every category, the pattern is the same: filters narrow what shoppers can find; recommendations show them what they didn't know they wanted.
Measuring Recommendation & Filter Performance
Track these metrics weekly to validate ecommerce product recommendation ROI:
For Recommendations
- Recommendation click-through rate (CTR), % of viewers who click a recommendation
- Recommendation conversion rate, % of clickers who add to cart
- AOV lift, current AOV vs baseline AOV before recommendations
- Revenue per visitor (RPV) from recommendation slots
- Cart abandonment rate with vs without recommendations
- Post-purchase upsell take rate, % of post-purchase offers accepted
For Filters
- Filter usage rate, % of collection page sessions using filters
- Filter-to-product conversion rate, % of filter users who view a product
- Filter-to-purchase conversion rate, % of filter users who buy
- Bounce rate on collection pages, with and without filters
- Average products viewed per session, should increase with good filters
- Mobile vs desktop filter usage, gap should be <20%
A recommendation engine that doesn't lift AOV by at least 10% within 60 days isn't working. Either the algorithm is wrong, the placement is wrong, or both.