Amazon - frequently bought together cross-sell If customers often buy a certain protective case for a type of headphones, then the Phone Number Database protective case is offered as “often sold together”. The beauty of this method is that almost all web shops already have this data. At Warmteservice, "often sold together" is also Phone Number Database the basis of the recommendations. The necessary statistics are Phone Number Database combinations of sold products. The more often combinations are sold, the higher the score. Consider the following tips to get started with data quickly: Base it on data you already register from customers and users.
You can quickly start Phone Number Database generating recommendations with this. An additional advantage of existing data is that you have already thought about the GDPR . Don't throw away dates. Many webshops do not store the clickstream data of users, while that data contains a lot of information. Keep at least the most Phone Number Database recent six months. Attempt 1: Frequently bought together The result with an initial basic approach based on sales data was encouraging, but not yet satisfactory. Installers (and many Phone Number Database other B2B customers) regularly buy the necessary materials for an entire job in one go. This can even be for a whole week or as a supplement to a small warehouse.
Generic accessories such as Phone Number Database couplings or general consumables, such as sandpaper and construction buckets, are regularly included. These items will therefore often appear as “sold together”, but are not necessarily a good cross-sell recommendation. Therefore, we have applied a correction. Analogous to Phone Number Database determining relevance in search results based on text. Frequently used words are given a lower weighting. Less common words are given a higher weighting (IDF – inverted document frequency ). Such a correction for general articles can also be useful for B2C, depending on the type of webshop and the purchasing behavior of customers.