Matching Engine

The Matching Engine is the heart of the Guided Selling Technology. It generates the product recommendations on the basis of customer requirements. Therefore it matches those customer requirements to all product-profiles that are available within a certain range of products.

Purchase decisions are usually situations of compromises, where product advantages and disadvantages need to be leveled with the user's highly-demanding requirements. The Product Advisor therefor can ascertain suitable alternatives, that mostly need the customer's wishes and avoids disappointing, empty result lists and lost sales.>

For the calculation of product recommendations, the matching engine analyzes customer profiles and compares them to all available product attributes. The Matching Engine falls back upon the recommendation strategies defined within the Knowledge Model to calculate the product recommendations.

• Of all the user provided input within one advisory session the advisor application will create a user profile that will be further refined within each of the advisory stages.

• A ranking of products is calculated according to how they match the user profile.

• At the beginning of the session a predefined selection of subjects and options is shown to the user, together with an initial result list optimized for the average customer

• Top recommendations (products that meet all consumer wishes and have a high selling probability) are located right at the top of the result list and are visually highlighted.

• If the products do not meet all of the customer's wishes, excentos generates alternatives that have the least deviation compared to the optimum, for example, products that are slightly more expansive or that have a similar color.

• Known product variants are handled correctly (no duplicates in the result list).

• Daily actualized push factors, such as stock, margin, top-selling products, that optimize sales results.

• For further sales enhancement, additional sorting rules can be implemented that positively influence the rank of recommendations in answer to the provided user input.  One example: “When „business lap top“ has been selected, show product X as the top-recommendation, followed by all items of product series Y.”

Top