Re-Do Digital
WEBINAR: Navigating the New Normal: Transforming Grocery Supply Chain with new Tech. Register Now

DeepRecs™: Visual AI
& NLP-powered Deep Recommendations

Personalize without behavioral data, with a human-like touch

Get a Demo

Algonomy is positioned as Visionary in Gartner’s Magic Quadrant for Personalization Engines

Our Customers

Leading retailers and brands across the globe have grown their revenue with Algonomy’s personalization software.

Overcome the Limitations of Traditional Recommendation Engines with Advanced AI Techniques

Surface Recommendations for Products Without Historical and Behavioral Data

Leverage your product data to contextually recommend new, seasonal, and long-tail products that would’ve otherwise remained buried due to a lack of historical events.

Replicate the In-store Experience on Your Online Store

Recommend products that are visually similar to and visually compatible with the product a shopper is looking at – the way a human sales associate would in a store.

Deliver Intuitive Shopping Experiences with Visual AI and NLP Deep Learning

Connect Products Based on Visual Similarity, Without the Need for Manual Tags

Leverage Visual AI and convolutional neural networks to detect and extract feature vectors and graph visual similarities between products. Generate relevant recommendations across fashion, lifestyle, and furniture verticals, and help shoppers make a decision.

Particle element

Deliver Complete-the-look Recommendations Based on Visual Compatibility

Apply Visual AI for ‘complete the look’ and other cross-sell strategies just like human merchandisers. Grow basket sizes without having the shoppers do the heavy lifting of finding matching products across categories.

Surface Related Products, Even for Fast-changing Catalogs

Use NLP-based deep learning algorithms to analyze catalog descriptions, reviews, and other textual data to infer relationships between products. Automatically recommend relevant new launches and long-tail items, without having to rely on historical events.

Particle element

Re-rank Recommendations Based on the Individual Shopper’s Intent

Combine deep learning with behavioral data to dramatically improve the quality of recommendations that reflect an individual shopper’s specific needs and preferences.

Request a Demo

Take Your Product Recommendations to the Next Level

Fill recommendation gaps and do away with fallback on top sellers.

Surface more of your catalog, including long-tail products.

Use product data — text data and product descriptions to form associations.

Let deep learning use subtle visual attributes to make recommendations, just like a human.

Draw Inspiration From Our Personalization Resources

Explore our eCommerce personalization resources — best practices, case studies, videos, and more — to stay ahead of the curve.

Other eCommerce Personalization Tools

Explore other solutions from our end-to-end personalization software, and learn how they enable individualized experiences at different shopper touchpoints.

Find™

Find™

Personalized Search

Deliver unique, contextual search results based on the individual shopper’s behavior. Eliminate instances of zero results with search that learns from the wisdom of crowds.

Discover™

Discover™

Personalized Browse and Navigation

Eliminate decision fatigue by presenting the most relevant products upfront and dynamically re-sorting category pages as per a shopper’s real-time intent.

Recommend™

Recommend™

Personalized Product Recommendations

Leverage 150+ pre-built strategies and a decisioning engine to curate 1:1 recommendations that boost conversions and basket sizes.

Social Proofing

Social Proofing

Live Activity Tracking and Badging

Increase engagement and conversion rates with urgency messaging based on real-time views, purchases, and inventory levels.

Deliver unique, contextual search results based on the individual shopper’s behavior. Eliminate instances of zero results with search that learns from the wisdom of crowds.

Learn More >

Eliminate decision fatigue by presenting the most relevant products upfront and dynamically re-sorting category pages as per a shopper’s real-time intent.

Learn More >

Leverage 150+ pre-built strategies and a decisioning engine to curate 1:1 recommendations that boost conversions and basket sizes.

Learn More >

Increase engagement and conversion rates with urgency messaging based on real-time views, purchases, and inventory levels.

Learn More >

What Leading Brands Say About Our eCommerce Personalization Software

“With Algonomy DeepRecs NLP, recommendations are based on product descriptions, rather than past purchases or historical browsing data. As a result, even for highly specialized and seasonal products, we can now recommend products with similar affinities, which makes shopping very convenient, highly relevant, and valuable for our shoppers.”
<span style="color: #7B7B7B !important;">CEO</span> <img src="https://algonomy.com/wp-content/uploads/2021/11/bdb843f6ad1341b42410-2.png" style="max-width: 240px; margin-top: 10px; display: block;" />
Katagiri Fumio
CEO

“Deep recommendations with NLP is right now the top strategy, and is delivering average attributable sales of Eur 10.68 per click. The results are scaringly good. Without Algonomy, we wouldn’t have used these innovative AI technologies that differentiate us, and help us grow.”
<span style="color: #7B7B7B !important;">Head of Ecommerce</span> <img src="https://algonomy.com/wp-content/uploads/2021/07/verkkokauppaa.png" style="max-width: 220px; margin-top: 5px;display:block;" />
Anton Paasi
Head of Ecommerce

“We instinctively knew that using visual aspects of a product for recommendations is effective in fashion and lifestyle business – it’s much closer to the expertise of our merchandisers. I am excited with early results – our engagement is up 40% over our merchandising rules, and revenue per 1000 impressions has increased by 19%, compared to the other recommendation models.”
Head of Omnichannel Customer Experience

“With Algonomy DeepRecs NLP, recommendations are based on product descriptions, rather than past purchases or historical browsing data. As a result, even for highly specialized and seasonal products, we can now recommend products with similar affinities, which makes shopping very convenient, highly relevant, and valuable for our shoppers.”
Katagiri Fumio
CEO

Simply amazing!

“Deep recommendations with NLP is right now the top strategy, and is delivering average attributable sales of Eur 10.68 per click. The results are scaringly good. Without Algonomy, we wouldn’t have used these innovative AI technologies that differentiate us, and help us grow.”
Anton Paasi
Head of eCommerce

Simply amazing!

“We instinctively knew that using visual aspects of a product for recommendations is effective in fashion and lifestyle business – it’s much closer to the expertise of our merchandisers. I am excited with early results – our engagement is up 40% over our merchandising rules, and revenue per 1000 impressions has increased by 19%, compared to the other recommendation models.”
Head of Omnichannel Customer Experience

Simply amazing!

Create Frictionless Experiences with Deep Recommendations

When a shopper visits an eCommerce store, their intent is to find what they are looking for in the shortest possible time.

Use Algonomy DeepRecs to improve product discovery and grow the number of returning customers to your online store.

Deepen Customer Loyalty with Human-like Product Recommendations

FAQs

DeepRecs helps businesses overcome key challenges such as:
  • Removing constraints associated with traditional recommendations that don’t work for scenarios with sparse data — seasonal and long-tail products, fast-changing catalogs.
  • Helping an online shopper who’s viewing a product find more products with similar and complementary visual features and attributes — just like a human sales associate would in a physical store.
Businesses have experienced a significant improvement in KPIs with DeepRecs:
  • 25X Revenue Per Million Impressions (RPMI)
  • 2X Engagement
  • +6.25% Average Order Value (AOV)
  • +4.99% Click Through Rate (CTR)
Yes. Through a simple user interface, merchandisers can set custom weights for different attributes such as brand, category, price, newness, and more. In addition, personalization delays can be set to impact how quickly you want to influence search results. Merchandisers can also set boost and bury rules to meet a retailer’s business, brand, or margin commitments.
The personalization software is equipped with an Experience Optimizer (XO) that selects the winning strategy based on the Unified User Profile Service (UPS) and the chosen KPI — RPV, AOV, Conversion Lift, etc. This AI-driven decisioning by XO ensures all strategies compete in real-time, and the highest performing strategy is chosen, such that the most relevant product/content is shown to every shopper for each interaction, without the need for manual merchandising and rules.
Algonomy understands the unique needs of different retail segments and has proven expertise with the verticals of fashion and apparel, jewelry, furniture, and more.
Anyone with product text descriptions is a candidate. Text data can be provided as a separate feed or within the catalog feed. NLP uses text fields, product name, brand, other attributes as well as customer reviews.
Leave the decisioning to AI - XO will decide the most suitable strategy for a placement based on the user profile, context, your goals, and availability of behavioral data.
Visual AI has been pre-trained on large image datasets as well as merchandiser curated looks. The neural networks extract feature vectors from the images and work with new catalogs with minimal incremental training. For complete-the-look, your merchandisers only need to define the categories to be recommended, and the algorithm uses images as well as behavioral data to tailor recommendations.

FAQs

DeepRecs helps businesses overcome key challenges such as:
  • Removing constraints associated with traditional recommendations that don’t work for scenarios with sparse data — seasonal and long-tail products, fast-changing catalogs.
  • Helping an online shopper who’s viewing a product find more products with similar and complementary visual features and attributes — just like a human sales associate would in a physical store.
Businesses have experienced a significant improvement in KPIs with DeepRecs:
  • 25X Revenue Per Million Impressions (RPMI)
  • 2X Engagement
  • +6.25% Average Order Value (AOV)
  • +4.99% Click Through Rate (CTR)
Yes. Through a simple user interface, merchandisers can set custom weights for different attributes such as brand, category, price, newness, and more. In addition, personalization delays can be set to impact how quickly you want to influence search results. Merchandisers can also set boost and bury rules to meet a retailer’s business, brand, or margin commitments.
The personalization software is equipped with an Experience Optimizer (XO) that selects the winning strategy based on the Unified User Profile Service (UPS) and the chosen KPI — RPV, AOV, Conversion Lift, etc. This AI-driven decisioning by XO ensures all strategies compete in real-time, and the highest performing strategy is chosen, such that the most relevant product/content is shown to every shopper for each interaction, without the need for manual merchandising and rules.
Algonomy understands the unique needs of different retail segments and has proven expertise with the verticals of fashion and apparel, jewelry, furniture, and more.
Anyone with product text descriptions is a candidate. Text data can be provided as a separate feed or within the catalog feed. NLP uses text fields, product name, brand, other attributes as well as customer reviews.
Leave the decisioning to AI - XO will decide the most suitable strategy for a placement based on the user profile, context, your goals, and availability of behavioral data.
Visual AI has been pre-trained on large image datasets as well as merchandiser curated looks. The neural networks extract feature vectors from the images and work with new catalogs with minimal incremental training. For complete-the-look, your merchandisers only need to define the categories to be recommended, and the algorithm uses images as well as behavioral data to tailor recommendations.