Blog

Join our Cropster Data Project!

created by Stefan P

In part 1 of our series on the Cropster Data Project we looked at the beginnings of the project and its goals: finding new relationships within the roasting data and using this knowledge to help roasters and the greater coffee community.

After launching the project, we contacted some of our customers in order to see if they were open to participate in research projects. While we hoped for the best, we weren’t prepared for the overwhelmingly positive feedback we received. Many of our good friends and customers were interested and gave us permission to analyze their data as part of the data project. We published a landing page for the project on our website and encouraged a handful of roasters with a scientific interest in participating to join. We are always looking for more companies to join the party and welcome everyone with open arms. As every customer is unique, there are always new insights to be discovered. If you are interested, please get in touch with us!

You might wonder what happens with our participant’s data. We take data security very seriously and ensure that the data is used only for the things that were agreed upon in advance. We only share data with selected university employees and associated students that we personally know and have a trustworthy relationship with. Furthermore, all researchers sign individual nondisclosure agreements (NDAs) that limit the use of the data to the scope of their individual projects. As an IT company, we also have technical solutions in place that hinder the sharing of sensitive data from the very start. We have developed a dedicated data pipeline that ensures sensitive data never leaves our system. We also use a software tool that serves as an interface between our database and our internal data team. It removes sensitive data such as company names, worker names and other private information by either hashing or removing sensitive data from the result set before it even leaves the database server. With this technique, we can ensure that research partners never see sensitive or personal data. They only see anonymized identifiers, that only we can map back to their true values.

Expanding the Coffee Cosmos

At Cropster we believe that innovation is key for any business and we developed a few ideas about how we could interpret the data in our database in new ways. In order to get fresh ideas from outside of our coffee cosmos, we reached out to different universities in Austria. Many of our engineers studied computer science in Innsbruck and it was a nice opportunity to get in touch with professors and former friends who became senior researchers to see if we could collaborate. This is how the data project came to be a collaborative project with experts inside and outside the specialty coffee industry.

In the first phase of the data project we approached the University of Innsbruck and drafted a proposal for a research project, in which we would analyze the relationship between roast curves and flavor labels. A talented student to support the senior researcher in conducting experiments and implementing prototypes was soon found. NDAs were signed and the first data project was ready to start!

In part 3 of our series on the Cropster Data Project we’ll take a look at the results of our research and the implications it has for the coffee community.