Personalized recommendations
Deliver personalized, relevant product and content recommendations
Provide tailored recommendations across email, web, mobile, and more as you execute your personalization strategy to optimize the customer experience (CX) and grow revenue. Oracle Recommendations suggests items to customers they are likely to desire but may not have found on their own. This, in turn, helps drive engagement and grow revenue.
Easily configure layouts and employ a single set of inventory items across channels (email, web, API) along with features, such as update schedules, map attributes, inventory profile, and error detection, all of which don’t require coding expertise. Oracle Recommendations then applies machine learning to surface the most relevant items (content or products) to each customer.
Types of recommendations
Personalize customer experiences using various algorithms behind the scenes to determine the best product or content to feature. Algorithm-driven recommendations can include:
Best sellers
Based on items that were most frequently bought.
Viewed this, viewed that
Based on items that were most frequently viewed together with the item being currently viewed.
Viewed this, bought that
Based on items that were most frequently bought by visitors who also viewed the currently viewed item.
Bought this, bought that
Based on items that were most frequently bought together with the currently viewed item.
Most viewed
Based on items that were most frequently viewed.
Last viewed
Based on the last item from someone’s prior visit.
One-to-one visitor affinity
Based on items a person is most predicted to interact with based on their interaction history.
Recommendations for personalized web experiences
Tailor website experiences through an integration with Oracle Maxymiser Testing and Optimization. Oracle Recommendations enhances features already available in Oracle Maxymiser, so you can easily drop in product or content recommendations to customize and improve each customer’s experience.
- Leverage preconfigured widgets, available algorithms, and an intuitive WYSIWYG editor.
- Create rules-driven personalization to various audiences through a drag-and-drop interface.
- Build light boxes, overlays, banners, or notifications and trigger them based on specific user actions.
For example, you can target by weather, so visitors from warm climates don’t see irrelevant content. By using geolocation and weather to target visitors, you ensure that people from cold climates see cold weather images. In contrast, visitors from warm climates see corresponding warm weather images.
Recommendations for personalized emails
Improve email click-through and conversion rates by adding algorithm-driven recommendations into Responsys Campaign Management. Select an algorithm from our library and use email open-time support to ensure that recommendations are relevant when viewed.
- Include product or service recommendations, natively, in Oracle Responsys emails.
- Improve the email click-through and conversion rates of session and cart abandonment emails by adding algorithm-driven recommendations consistent with those on the website.
- Open-time support ensures recommendations are timely and don’t frustrate the recipient (for example, making a recommendation of an out-of-stock item).
Recommendations with other applications
- Use recommendations across any customer-facing system—not just Oracle CX solutions.
- Make recommendations on mobile, third-party applications.
- Use a REST API service so your developers can rapidly extend recommendations to mobile, IoT, and beyond.
Recommendations in use
Anonymous shoppers (B2C, B2B2C)
Problem: Retailers and brands must appeal to new and returning customers looking for either the "latest and greatest" or a specific item based on their preferences and past purchases.
Solution: Surface the most popular, best-selling, or trending items.
Benefit: Raise conversion rates and average order values.
Retargeting (B2C, B2B2C)
Problem: Retailers, brands, and media platforms selling directly to existing consumers must make every interaction personal and relevant to boost retention and maximize loyalty.
Solution: Leverage rich profile insights to inform personalized, contextualized recommendations.
Benefit: Grow average order value, shopping frequency, and customer lifetime value.
Long sales cycle services (B2B, B2C, B2B2C)
Problem: Long sales cycle services have infrequent, often anonymous, shoppers who visit brands' websites multiple times as they look to compare prices and solutions.
Solution: Based on independent attributes, low context, and limited data, use the "last viewed" algorithm-based recommendation model to allow visitors to pick up where they left off.
Benefit: Minimize frustration for complex exploration and buying journeys to maximize the chance of a conversion.
Next steps
Purchase Oracle Recommendations as a stand-alone solution to integrate with your other martech applications—Oracle and non-Oracle—or purchase it as part of Oracle Infinity Behavioral Intelligence.
Learn more about Oracle Infinity