How many times have you entered a store, physical or online, looking to buy a gift or something for yourself and, despite hundreds or thousands of products, saw nothing interesting and left disappointed? Or else, due to lack of time, did you buy a product that wasn't quite what you were looking for and later regretted it? Sometimes this happens because the stores don't actually have what you were looking for. In other cases, however, they just didn't know what to offer you.
As we already know, companies all over the world have chosen to diversify their portfolios and offer more options to attract consumers who previously did not buy their products, or to increase the average ticket of their usual audiences. The big problem is that, many times, the huge offer of products can cause confusion in the consumer, causing him to give up the purchase, mainly in the virtual environment, where it is possible to visit dozens of different stores with a few clicks.
Figure 1 – So many options, so many products… what to choose?
Source: Netshoes, Submarino, Americanas.com, Saraiva
Aware of this problem, stores and service providers have sought to direct part of their portfolio to each customer, in order to provide a better shopping experience, developing product recommendation systems based on the buyer's profile. One of the most emblematic examples is that of the streaming giant Netflix, which sees its recommendation system as an important pillar for user satisfaction.
The company has the challenge of offering a wide variety of titles to its subscribers, but each content offered has a licensing maintenance cost (think of it as an inventory cost). Therefore, adding more and more movies and series to please everyone simply because of the high volume can be detrimental to the profitability of the business. It is more advantageous to efficiently target a leaner portfolio. The company also measures the performance of its recommendation system through indicators that measure the acceptance of suggestions (called take rate) and the number of relevant titles as a function of the total available (the Effective Catalog Size). Your former vice president of innovations posted a article in 2016, in which he further describes the highly personalized system, its metrics and algorithms, and how it helps Netflix save more than $1 billion a year by reducing the need for investments to acquire new customers.
Figure 2 – Example of a custom Netflix interface
Source: Netflix
The example of Netflix can very well propose reflections to see what your company is doing to boost the sale of its entire portfolio, and generate more value for customers, with product suggestions that increase their satisfaction. As the streaming giant does, a recommendation system can be used to reduce the need to have numerous SKUs available or even to increase the turnover of your products, reducing waste and improving customer service and loyalty.
Here is our recommendation.
References
Carlos A. Gomez-Uribe and Neil Hunt. 2015. The Netflix recommender system: Algorithms, business value, and innovation. ACM Trans. Manage. Inf. Syst. 6, 4, Article 13 (December 2015), 19 pages.
Fernando Chalréo
He has been working on consulting projects in Logistics and Supply Chain for 5 years, with experience in companies in the consumer goods, retail and food and beverage sectors. Types of projects already carried out: Sales & Operations Planning, Inventory Management, Network Planning, Business Process Review, Logistics Indicators and Transport Management
