Product recommendation systems powered by deep learning provide highly relevant recommendations to each customer using real-time user data to make accurate recommendations.
But eCommerce teams can’t always let the artificial intelligence-powered algorithm run wild, completely unrestricted.
To control the AI, you can add flexible business controls or ‘recommendation rules’ to the machine learning models that power your recommendation engine. They add the human touch to enable real-world personalization. Depending on how it’s set up, you can finetune your recommendation strategies so that the product selections are more closely aligned with campaigns, promotions, and wider business goals. These controls effectively enable filtering and act as ‘AI Overrides’ by incorporating a business logic layer overlaid on top of the
We’ll now try to understand when is it best to incorporate business controls – including promotion, blacklists, and conditional rules. But before that let’s see why it’s advisable to start the recommendation algorithm un-ruled.
Start Recommendation Algorithms As Un-Ruled, Layer In Rules Over Time.
Product recommendation engines powered by deep learning are constantly studying user behavior in real-time. By applying multiple complex rules to the algorithm, you may hamper its reinforcement learning process. Plan your recommendations strategy with only the most necessary rules to start. Once the recommender algorithm has had enough learning and begun to understand user preference, you can start adding rules iteratively. It’s always a good practice to be sure of the effects of the change on existing rules before incorporating any additional rules.
When eCommerce Teams Should Implement Rules, No Questions Asked
When there is a contractual obligation.
It’s commonplace for multi-brand retailers, general merchandisers and department stores to have restrictions on which brands they can merchandise side-by-side and which they cannot –
-i.e. competitor brands that cannot be positioned against each other. In this instance, a conditional rule would be applied. For example – if User A is viewing ‘X’ brand product, your personalized product recommendations cannot show brand ‘Y’ products.
When the product catalog includes non-revenue items such as samples.
Some retailers have non-revenue items in their product catalog that are only to be shown on certain pages or under certain conditions. For example, beauty brands leverage samples to get visitors deeper into their product catalog as a risk-free trial. In this instance, you can set up a rule to blacklist sample products from certain carousels or only restrict them to certain pages.
Examples Of Product Recommendation Algorithm Rules
When there are explicit product pairs
Certain products have defined product pairings or complementary product offerings that can be shown simultaneously as a recommended product to increase your chances of cross-selling. An example is if a visitor is viewing a bike, then the personalized recommendation could be a maintenance kit. In these instances, an override carousel slot would be used to force the maintenance kit to be shown when the visitor is viewing a bike.
When the business wants to promote certain categories.
To best align to business goals and targets such as the launch of a new or seasonal collection, the ‘promote’ rule can be leveraged. This increases the likelihood for products within that category to show up to visitors.
When the category shown should be restricted.
Certain categories are retailers’ bread and butter and if a visitor has landed on one of those pages and there is only one recommendation carousel, it’s best not to direct their on-site journey, elsewhere. For example, if a visitor is on a bags page of a luxury retailer, recommendations should be filtered on similarity to exclusively show other similar products using the ‘only show’ rule.
When price banding should be implemented.
Price banding can be leveraged in a couple of ways as aligned with wider business goals including on the PDP and on the basket page. On the PDP, merchandisers can leverage rules to only recommend products that are within a certain price range of the product being viewed. On the basket page, rules can be implemented to recommend products to get visitors above a certain threshold and increase the value of their basket.
Recommendation rules are powerful mechanisms for eCommerce teams to add an element of curation to product recommendations. They add personalization and can elevate your customer experience. But more importantly, they help you align your recommendation strategy with business needs or goals. We always recommend starting the machine learning algorithm as “un-ruled” as possible, then layer on rules over time. Doing so sets the baseline and maximizes performance. The last thing eCommerce teams want is to be 6 months in leveraging deep learning recommendations and discover that a rule based on an early assumption was costing 0.5% RPV.
At Coveo Qubit we are data-driven and use data science to be laser-focused on achieving high success rates for our customers. Generic solutions that treat every customer the same breed generic results. That’s why we’re always at the forefront of new technologies that enable us to drive a truly personalized eCommerce experience. We’re here to make eCommerce better.