Approach

Several types of recommendations are available. The users can configure few parameters and get those recommendations computed.

The different types of product recommendations available in the Product Recommendation Accelerator include frequently purchased items, products bought by similar customers, products bought together, similar products, similar products from another brand, up-sell, and down-sell recommendations.

The Product Similarity Accelerator can be leveraged in conjunction with the product recommendations provided by the Optimization - Product Recommendation Accelerator to enhance the accuracy and relevance of the recommendations. This allows users to provide more personalized and targeted product suggestions based on similarities between products, ultimately improving the effectiveness of the recommendations in sales scenarios.

Limitations

The model should be updated regularly, for example, once a month, to ensure accurate product recommendations based on past transaction patterns. The out-of-the-box integration within Quotes provides recommendations for the entire quote, and there are also code examples available for recommendations per item.

However, for "Up-sell" and "Down-sell" recommendations, the assessment of margin and price is not performed for the specific quote context but in average, which may lead to some discrepancies, requiring additional checking within Quotes.