The Smeets & Graas webshop sells supplements. Code050 have development (and maintains) this Laravel platform.
Many e-commerce platforms are not gathering data in a good fashion and even less are using their data in a correct way or at all. The goal of this project is to build a data-driven product recommendation system.
Gathering and storing data
Many websites only use a simple tool that scrapes standard data from the website and the user information. In this project, we extensively use google tag manager and google analytics to optimize the data gathering process, using all the data that may legally be stored from users and transaction data. One of the problems in this phase is linking the correct transaction or non-transaction to the visit/session. Storing the data is another issue that many platform owners underestimate. Only storing data in Google analytics makes it difficult to work correctly with the data.
Furthermore, the data in Google Analytics is technically not under your ownership. Since we live in a data age, we ensure that the information is correctly copied and added to your database. This gives you the excellent functionality of existing scraping tools combined with the ownership of the data.
We are building a recommendation system.
When the data is gathered and stored properly, data science and AI can go wild to give you amazing predictions and recommendations. For example, with linked session-level data and transactions, we can predict whether visits end without or with transactions (product being bought or not). Furthermore, with good storing of the products purchased per transaction, we can use the robust Apriori algorithm to give data-driven product recommendations. This means that users will get the best recommendation based on your data!