Make Ecommerce Recommendations A Team Multiplier
Maximize revenue with ecommerce recommendations that work with, not against, your product discovery platform.
Ecommerce Product Recommendations That Work For Your Bottom Line
Constructor’s recommendations give you the power to optimize for any ecommerce business KPI — unit sales, AOV, margins, or anything else. Recommendations are constantly validated against observed click and conversion behavior to determine the right products to show users at the right time to guarantee increases in the outcomes that matter most.
Personalize With A Deep Understanding of Your Users
Constructor informs recommendations with everything known about a particular user — as well as other similar customer histories — to present highly-personalized, KPI-optimized recommendations that create better ecommerce experiences and drive more business value, in under 100ms.
Recommendations That Work With, Not Against Your Search
With Constructor, recommendations are tied into every other facet of your product discovery. This means a dollar made in recommendations isn’t a dollar lost in search or browse.
Don’t Just Recommend The Popular Products - Recommend the Right Products
Older recommendation systems for ecommerce only recommend products with lots of data— the popular ones. With Constructor’s embeddings-based approach, the right products get recommended for every user, even when they’re new and don’t have a lot of data.
How it Works
- Recently viewed products
- Alternative products
- Complimentary products
- User-featured products
The simplest but often most-interacted-with recommendation type: Show a user’s most recently viewed items.
Show products a user might consider as an alternative to a particular product or set of products. For example, if a user is looking at toothpaste, this algorithm shows other types of toothpaste.
Alternative items are most relevant when a customer is deciding between several potential items of a similar type. Imagine you’re viewing a product detail page for hot dog buns. Alternative recommendations would likely include other hot dog buns you might consider as an alternative to the product you’re viewing.
Show products a user would purchase in addition to a particular product or set of products. For example, if a user is looking at toothpaste, this algorithm shows toothbrushes and dental floss.
Complementary items are most relevant when a user has demonstrated interest in purchasing a particular item, capitalizing on this intent by suggesting items to purchase in addition to the particular item. Users adding chips to their cart might want guacamole and salsa, while users with hot dog buns in their cart likely want ketchup and hot dogs.
Provide personalized recommendations to a user based on their entire view and purchase history. This algorithm is valuable on homepages and zero result pages — anywhere outside the context of a particular product page.
Simply showing the items most often viewed or bought with a particular item is table stakes these days. More challenging (and crucial) is coalescing a user’s entire view, click, add-to-cart and previous purchase history across multiple sessions from weeks to seconds ago and providing individualized recommendations.
Don’t just trust us. Make us prove it.
Let us quantify the value of Constructor’s ML-backed search and discovery on your site using your data. No contract required.