Project description

Creating a new interface for optimizing customers’ inventory. By using sales history and machine learning the system is able to predict future sales and give users a better suggestion on what to order and when. By adding graphs to visualize the outcome depending on what parameters the user enter, managing inventory will become more efficient and easy.

Role and responisibilites

In this project I collaborated with another designer in creating wireframes but I was responsible for making the clickable prototype.
In June 2023 we presented the MVP version of the ‘Inventory optimization’ prototype to the development team and a few users. The feedback on the feature itself as well as the interface was very positive and something that became a nice show-off feature for Visma Net when presented to partners.

Work process

The project started in 2021 with a design sprint where a prototype was created and tested with users. The project was then put on hold until 2022. When the project started again it was decided that the work from the design sprint was not sufficient. More research was needed for the interface while the backend development could start in advance. The research was mainly done during late 2022/early 2023 by a UX researcher.
My role as a UX designer was to create wireframes with the help of research output and insights. I worked together with the research team and another designer to make sure information was displayed correctly and that the visualizations were changed accordingly to the parameters that user entered.

Tools used

Figma for designing

Challenges

A big challenge when designing an interface that uses machine learning is to make sure the users feel secure about the outcome. The design needs to present information in a way that users understand where it comes from. The design also needs to balance between making the work easier for the user while still letting them be in control if they need to.