Talk to our team

AI Recommendations

AI that shows shoppers first what they'll buy next

Reads shopper behavior and product context in real time, then suggests the products each shopper is most likely to buy now.

Talk to our team See customer stories
AI Recommendations screen: Picked for you

Changing recommendations changed both revenue and operations

In some cases, average daily revenue through recommendations grew 10 to 20 times over pre-launch levels, and productivity per placement grew 4 times.

kt alpha shopping
W Shopping
Shinsegae Live Shopping
Shinsegae Duty Free
See customer stories

Bestsellers aren't the answer for every shopper

Swap fixed displays for personalized recommendations and turn the next moment of browsing into a purchase.

Conventional recommendations: the same products shown to every shopper

Where one-size-fits-all recommendations fall short

  • Fixed placements driven by popularity
  • No basis for recommending to new shoppers or new products
  • Manual pinning and excluding add to merchandisers' workload
  • Hard to tie recommendation performance to revenue
A Plus AI recommendations: results that change for each shopper

Personalization that learns from behavior

  • Reflects real-time behavior and long-term preferences together
  • Recommends for new shoppers and new products too
  • Business rules set separately, the model stays as is
  • Performance measured from impressions to revenue

Personalize the key placements of the shopper journey with one engine

Run recommendations with different goals on the home page, product pages, order confirmation, and promotions from one engine.

Home personalization

Products that fit from the very first screen

Combines the context of each visit with past behavior to give shoppers a reason to start browsing. Because it reflects both what they just viewed and long-standing tastes, returning shoppers spend less time searching.

Personalized home feed for each shopper

Related recommendations

Products they're likely to view together

Learns from what was viewed alongside each product to suggest similar items, items to buy together, and alternatives. When a stockout or price would send a shopper away, an alternative keeps them in the store.

Related recommendations on a product page

Order confirmation recommendations

The next thing to browse, right after purchase

Shows shoppers who just placed an order personalized recommendations based on their behavior so far. A flow that used to end at checkout leads into the next round of browsing and a return visit.

Recommendations after purchase

Promotion personalization

Different promotions for different shoppers

Analyzes the products in each running promotion and surfaces the ones this shopper is likely to care about on the home page. Beyond promotions, any product pool with data can be personalized.

Personalized promotion recommendations

Better recommendations and more efficient operations, together

From model performance to running live placements and measuring results, we work closely with your commerce team.

DOMAIN-SPECIFIC

Built for commerce

  • Optimized for product discovery and purchase conversion
  • Shows the products most likely to be bought first
  • Factors in product name, brand, category, price, and more
BEHAVIOR-DRIVEN

Learns from shopper behavior

  • Learns from views, clicks, wishlists, add-to-carts, and purchases
  • Session signals apply instantly, long-term preferences through daily training
  • Handles new shoppers and new products with cold-start methods
RULE + AI

Combined with your business rules

  • Can exclude purchased items and items similar to them
  • Change a policy in settings and it applies right away, with no retraining
  • Less of the manual pinning and excluding merchandisers used to do
PROVEN IN PRODUCTION

Proven in production

  • Running at multiple commerce customers
  • +17% revenue in a pilot against an existing commercial recommendation engine
  • Every AI Recommendations pilot customer went on to full implementation

Frequently asked questions

Find answers to common questions before you get started.

It learns from shopper behavior, such as products viewed, added to cart, and purchased, together with each product's category, attributes, and price context, then picks the products most likely to be bought now. Each placement, such as the home page, product pages, order confirmation, and promotions, gets recommendations suited to its purpose, with business rules like stockouts and promotions applied too.

See it with your own products

Use your real data to see what changes before and after AI Recommendations.

Talk to our team See customer stories