May 10, 2021 • 2 MIN READ
At Ayazona, we try to be as scientific as possible about how we build our products. Our engineers generate hypotheses that we test by running experiments — generally in the form of an A/B test — to learn what works and what doesn't. The learnings give us insights and fuel new product ideas.
Recommendation algorithms are steadily taking centre at the core of our products. They provide our support team and customers with personalized suggestions to help build the most nutritious meal kits while reducing the amount of time and frustration to find the right items to include in a weekly meal & essentials kit when working with a minimal budget.
A recommendation engine is a system that suggests products, services, and information to customers based on data analysis. This data can be about the variety of products we have on our shelves, featuring product nutritional information and pricing, or it can be data about our customer's purchase patterns or both.
We see the recommendation engine playing a significant role in helping us realize our company vision given the unknowns of the problem we are solving while helping boost our impact and other essential metrics. In addition, it can have positive effects on our user experience, thus translating to higher customer satisfaction and retention.
Let's take an example of our weekly groceries and essential kits. Instead of having our team browse through thousands of food items and household products, Ayazona presents them with a much narrower selection of items that are likely to meet each customer's portfolio based on their weekly or monthly budget. This capability saves us time while delivering a better user experience to our customers. With this function, Ayazona impacts more low-income households while improving operations efficiency.
While our recommendation engine offers promising capabilities for the problems we are solving in our markets, there is still a long way to go until we see its full capabilities, with the sparsity of data as the critical challenge we currently face.
Thus, we continue to improve our algorithms and look for new areas to personalize while we understand our data catalogues. In addition, we intend to extend beyond the items selection layer by looking for new ways to use these systems to offer an integrated service to our customers.
We have spent the last four months pioneering a new recommendation engine at Ayazona. The recommendation engine is an innovative approach to extend the capabilities of Ayazona while remaining a unique product that significantly impacts our customers. However, we still feel it's early for experimentation at Ayazona.
We are constantly evolving our production practices. If you would like to know more, or if you're interested in joining the team and contribute to our journey, do not hesitate to reach out.
More from Engineering