Recombee is an AI-powered recommender that offers both recommendations & personalized search. Whether it’s recommending products, articles, or movies, Recombee leverages advanced algorithms and machine learning to deliver personalized recommendations & search that keep users engaged and satisfied. With its easy integration and customizable features, Recombee empowers businesses to optimize content discovery, increase user retention, and drive conversions effortlessly.
Highly-performing recommendations
Recombee prides itself in increasing conversion rates and in turn revenue tenfold. You can verify this by reading their case studies. Using the case studies, you can observe that Recombee improves website performance in most situations whether it is general purposes, e-commerce, streaming, news, etc…
Here are the main types of recommendations :
1. Item recommendations
This is the most common type of recommendation. It is typically used for general purpose websites, e-commerce, social media, etc… It consists of recommending content (pages, products, posts, etc…) to users.
There are two major forms of this recommendation available with Recombee : ‘items to user’ and ‘items to item’ (plus ‘recommend next items’ only used for infinite scroll and pagination).
2. User recommendations
This form of recommendation is most commonly used for social media. It will recommend users to users.
The two user recommendation types available with Recombee are ‘users to users’ and ‘users to item’.
3. Item segmentations recommendations
Item Segmentation provides a convenient way to organize items within your catalog by grouping them into segments based on their properties. These segments, such as categories, brands, genres, or other custom criteria, can then be recommended to users. For example, in e-commerce, items may be segmented by category or brand, while in media streaming, they may be segmented by genre or artist. Users can be recommended top segments based on their preferences. Additionally, Item Segmentation allows for exploring related segments to specific items, such as brands related to a product or artists related to a song. Users can also search within segments to find specific brands, artists, or other criteria.
2-in-1 solution : Also includes personalized search
Recombee also offers personalized search at no further cost.
Complete and intuitive dashboard
Recombee offers a complete and intuitive dashboard. From this dashboard, you can :
- use the KPI (Key Performance Indicator) page to view your recommendations’ performance using the detailed analytics.
- view all kinds of data (items, users, segmentations and series).
- create Recombee scenarios (full configurations to be used by recommendations) and set business rules for these recommendations.
This dashboard has been designed to be as easy to use as possible. From this, any beginner can evaluate the recommendations’ performance and the added value, view the data that has been sent and configure recommendations using scenarios. Recombee even offers HTML Widgets that can be configured from this dashboard using the editor and are the easiest way to get recommendations and full-text search into your site.
1. Real-time analytics (KPI)
From this page, you can determine how effective your recommendations are doing by viewing the Recombee charts.
Down below are all of the metrics you could display in the Recombee KPI charts :
- Recommendations
- Views
- Bookmarks
- Cart additions
- Ratings
- View portions
- Purchases
- Views from recommendations (both reported and estimated)
- Bookmarks from recommendations (both reported and estimated)
- Cart additions from recommendations (both reported and estimated)
- Ratings from recommendations (both reported and estimated)
- View portions from recommendations (both reported and estimated)
- Purchases from recommendations (both reported and estimated)
If you wish to learn more about the difference between estimated and reported success, you could read about it in the official Recombee documentation.
You could also view the success of each scenario individually using the following chart :
Or, you could check on the recommendations’ success rate overtime using the click-through rate and conversion rate chart :
2. View data
This page allows the user to view the data that has been added to Recombee. This is a screenshot of the items list but you could also view users, segmentations and series.
3. Full customization of recommendations using Recombee scenarios
Each Recombee scenario includes :
- the model used (in this case ‘e-commerce cross-sell’ model). You can check out a list of all the available Recombee models.
- filters : Filters are essentially business rules that determine which items will be shown to users, filtering them based on specific criteria.
- boosters : Boosters, like filters, are business rules that influence the items returned to users. However, boosters are less strict than filters. Instead of excluding items that don’t meet certain criteria, boosters prioritize elements that do, giving them preference in the recommendations.
- constraints : Constraints will determine how many times items of same categories or that share attributes will appear in one recommendation.
- widgets : you can customize each recommendations’ appearance using the HTML widget editor right from the Recombee dashboard.
- analytics : You can view detailed analytics for this specific scenario.
The filters and boosters business rules can be set either graphically (Recombee even includes a few default ones) but they cab also be set using ReQL : Recombee’s own query language. ReQL is a powerful tool that offers total freedom when it comes to customizing your recommendations and personalized search experience. You can learn more about this in the official ReQL documentation.
Easy integration : All the necessary methods and tools available
Recombee can be integrated into any and all websites thanks to all it’s methods, models, tools, etc…
1. API clients
Recombee offers quite a few API clients for different programming languages so you can always make full use of Recombee no matter the one you use. Recombee’s clients are fairly easy to use and fully documented.
If you are interested in deploying Recombee in a PHP application, you could check out our guide on how to easily setup Recombee with the PHP API client. It will allow you to test out Recombee’s performance and features for yourself in less than 30 minutes using a dataset containing 40k posts.
2. User events
Recombee makes use of collaborative filtering. Collaborative filtering uses machine learning to learn all about the users as a group. To do this, it needs to gather user events, which are relevant data used to train the models. Whenever a user interacts with the website in any way, a user event is sent to Recombee.
Recombee offers a wide range of events/interactions so that it can learn and know which item to send in any situation. The list of Recombee interactions are :
- detailed view of an item
- purchase of an item
- rating of an item
- item added to cart
- item bookmarked
- viewed portion of an item
3. Models
Recombee offers many models. You can check all of them out in the official documentation.
With 32 different models available for both recommendations and personalized search, Recombee is a very versatile recommendation engine. They have general purpose models but also models tailored to personalized search, e-commerce, media streaming and advertising (recommend similar items, popupular items, visual similarity, etc…). This means that no matter the website or service, Recombee will be the right fit for you.
Pricing
You could check out our comparison of different personalization solutions based on price.
How to add Recombee to WordPress & Woocommerce
Using WPSolr, a WordPress search & recommendations plugin, you can easily set up personalized search and recommendations powered by Recombee with no code needed.
Recombee being a recommender designed for e-commerce, you could integrate Recombee into Woocommerce using WPSolr.
WPSolr is also compatible with a host of other search & recommendations engines (Apache Solr, Elasticsearch, Opensearch, Algolia, Weaviate, etc…) so you can fulfill all your needs and greatly increase your CTR with one single plugin.
You could check out our WordPress recommendations plugin comparison.
