When was the last time your product recommendations delivered a breakthrough result?
Product recommendations have become essential table stakes in ecommerce — and for good reason. They work. But after more than a decade of relying on the same foundational machine learning techniques, even well-optimized recommendation strategies have reached a performance plateau.
The challenge isn’t that traditional recommendations are failing. It’s that they’ve hit their ceiling. When everyone is using the same approaches, even the best-performing “Personalized for You” carousels deliver similar results across the industry.
However, recent breakthroughs in AI technology are raising that ceiling. The brands that are leveraging these advancements are seeing results that push beyond the plateau: double-digit lifts in click-through rates, conversions that surge by 35%, and revenue increases that translate to millions in additional annual impact.
Here’s how Bloomreach’s Recommendations+ is delivering these results across three critical touchpoints — and why this represents a significant evolution.
What Makes Recommendations+ Different: The Technology Behind the Results
In the early days of ecommerce, machine learning techniques like collaborative filtering and matrix factorization were cutting-edge ways to personalize product recommendations. However, that first generation of personalization soon became widely adopted, and these types of recommendations serve as the baseline for any ecommerce experience.
More recently, though, breakthrough technologies like neural networks and transformers (aka the “T” in ChatGPT) have once again changed how AI can understand and predict customer behavior. These types of advancements are what set Recommendations+ apart from traditional recommendation engines.
At its core, Recommendations+ is powered by a Sequential 2 Tower Neural Network (2TNN) — the same predictive AI modeling technology that powers transformers. This represents commercial access to completely new AI approaches that simply weren’t available in ecommerce platforms until now.
This breakthrough technology enables three core capabilities that drive measurable results:
Sequential Learning
Recommendations+ understands each customer’s unique journey, analyzing their browsing path and evolving intent to predict the next best product most likely to foster engagement. It’s not just looking at what they’ve clicked — it’s understanding the sequence of their behavior to anticipate what comes next.
Real-Time Personalization
Immediate learning and adaptation mean more relevant recommendations for first-time and anonymous visitors. Different products are showcased with each subsequent click, learning and changing after every single interaction. This isn’t batch processing — it’s instantaneous optimization.
Historic and In-Session Behavior Analysis
For loyal customers, Recommendations+ analyzes their purchase and product history while simultaneously capturing complex, non-linear user behavior patterns in the current session. This enables deeper, more granular personalization based on both long-term preferences and immediate shopping intent.
Next, let’s look at some use cases that demonstrate how Recommendations+ can drive engagement throughout the customer journey.
Use Case #1: Driving Traffic Through Email Newsletters
Weekly and recurring batch campaigns are designed to do one thing: drive traffic from email to your website. But standard recommendations weren’t maximizing engagement potential, leaving marketing teams with passive email readers instead of active site visitors.
A large furniture retailer faced exactly this challenge with its weekly newsletter campaigns. The brand needed to convert subscribers into engaged shoppers, and product recommendations were the natural vehicle — but the existing approach wasn’t delivering great results.
The Implementation
The retailer implemented Recommendations+ in weekly newsletter campaigns with a straightforward A/B test: 50% of its audience received Recommendations+ recommendations, while 50% continued with standard recommendations.
The key difference? Recommendations+ was optimized specifically for click-through — the critical metric for newsletter success.
The Results
Over a two-month test period, the results were striking:
- 6.5% increase in conversion rate
- 16% increase in revenue
When projected across a full year and deployed to 100% of traffic, this translates to potentially millions in additional revenue — all from a single weekly email campaign.
Why It Works for Newsletters
Newsletters are designed to drive traffic to the website, and Recommendations+ optimizes specifically for clicks. More clicks lead to more product views and ultimately more conversions, creating perfect alignment between the channel goal and Recommendations+ optimization.
The sequential learning capability means Recommendations+ understands not only what products are popular, but also which products are most likely to inspire that critical click-through based on each subscriber’s unique preferences and browsing history.
Use Case #2: Reengaging Customers With Abandoned Cart Emails
Abandoned cart emails are one of the most critical customer recovery touchpoints. But they need to accomplish two things simultaneously: bring customers back to complete their purchase AND suggest relevant additional products that might inspire them to expand their basket.
The same large furniture retailer recognized that its abandoned cart flows weren’t maximizing this dual opportunity. It needed to balance cart reminders with product discovery in a way that felt personalized, not generic.
The Implementation
The retailer integrated Recommendations+ personalized recommendations into its abandoned cart flows, running an A/B test alongside a standard recommendation strategy. The real-time personalization adapted to each customer’s cart contents and browsing history, creating contextually relevant suggestions.
The Results
The impact on abandoned cart recovery was immediate and substantial:
- 35% increase in email click-through rate
- 32% increase in conversion rate
Why It Works for Abandoned Carts
Abandoned cart emails capture a moment of high intent, and Recommendations+ maximizes that opportunity by understanding the customer’s immediate shopping context. Sequential learning analyzes where the customer was in their journey when they abandoned, while real-time personalization recommends complementary products based on cart contents.
This increases both cart recovery and basket expansion opportunities. A customer who abandoned a sofa might be shown complementary accent chairs or coffee tables that fit their browsing patterns — products they genuinely want to see, not random suggestions.
Use Case #3: Deepening Engagement on Product Detail Pages
Here’s a question that challenges conventional wisdom: If customers are already engaged on a product detail page, why try to drive more clicks? Doesn’t that risk pulling them away from a potential purchase?
A major jewelry retailer asked this exact question. The concern was valid — PDPs represent the mid-funnel moment where customers are closest to conversion. But the jewelry retailer hypothesized that the right recommendations, shown to the right customers, could deepen engagement rather than distract from it.
The Implementation
The brand implemented Recommendations+ on product detail pages in its US market, adding a “Personalized for You” widget below product details. The key strategic decision was targeting this specifically to returning visitors with browsing history, ensuring they would see truly personalized recommendations instead of generic cross-sells.
The brand A/B tested this against its standard recommendations to measure the true incremental impact.
The Results
Over a one-month test period, the results validated the brand’s hypothesis:
- 26% lift in conversion rate
- 35% lift in total revenue
When projected across the US market for a full year and deployed to 100% of traffic, this translates to €1.5M in additional revenue — from a single widget placement.
Why It Works Even for Mid-Funnel Journeys
Even on PDPs, continued engagement is valuable. Recommendations+ helps customers discover better-fit products by understanding the browsing journey that led them to this specific page.
Sequential learning recognizes patterns — perhaps they’ve been comparing similar items, or maybe they’ve been browsing complementary products. Real-time adaptation ensures that different products are shown with each click, keeping the discovery process fresh.
Here’s the insight: Customers who explore multiple products often convert at higher rates. By facilitating smarter exploration rather than random browsing, Recommendations+ increases both the likelihood of conversion and the average order value.
The Common Thread: How Recommendations+ Drives Results Across Use Cases
These use cases — newsletter campaigns, abandoned cart emails, and product detail pages — represent very different moments in the customer journey. And yet, Recommendations+ delivers measurable impact across all of them. Here’s why:
Optimization for Engagement First
Recommendations+ primarily optimizes for product click-through rate (in addition to add to carts and purchases), operating on a simple but powerful principle: more clicks lead to more views, which lead to more conversions. This is particularly powerful for driving traffic from external channels like email and mobile, where the initial click represents the critical gateway to engagement.
Works for Both New and Returning Customers
Anonymous and first-time visitors benefit from real-time learning that creates relevance from minimal data. Returning customers experience deep personalization powered by historic behavior analysis. This dual capability means Recommendations+ improves both first-session conversion rates and long-term customer lifetime value.
Zero Technical Burden
There’s no special technical setup or catalog requirements. Recommendations+ works with standard ecommerce catalogs and offers a visual interface that non-technical marketers can use to deploy recommendations across any channel quickly. The AI sophistication is completely abstracted away from the user experience.
Embrace the New Standard for Product Recommendations
The business impact of Recommendations+ is tangible:
- Double-digit improvements in CTR, conversion rates, and revenue
- Fast implementation with zero technical burden
- Performance gains for both new and returning customers
- Applicability across critical ecommerce touchpoints — email, abandoned cart flows, and on-site experiences
For marketing teams seeking to differentiate their customer experience, and for brands wanting to maximize the ROI of their recommendation strategy, Recommendations+ offers a proven path forward. Ready to see what Recommendations+ can do for your business? Learn more about Recommendations+ or contact your Account Manager to see it in action.
