We believe in the power of mathematics. Numbers rule our world, they define the laws of nature, and they – and this is crucial – represent facts rather than opinions. Numbers are therefore the essential building blocks of our work, the driver behind our success.
We understand that not everyone’s a data geek, so we do our best to transform our world of numbers into the world of meaningful sentences. Because that’s how people communicate and interact with each other. So the way we write our code and the way we organize our time in development is to make the transition between those two worlds as seamless as possible. While we are not perfect yet and may not perfect all our methods for years to come, we still see this as an advantage.
We know what needs to be done to find this little singularity of ours – defining our work path and taking so much time of our weekends and after-hours.
From the start ( 2004) we have focused mainly on the data part of our job, and we believe we have found the perfect balance between the old world of classical algorithms and the modern era of artificial intelligence. We believe that using the best of both worlds is most meaningful and produces the best results for our customers.
Our architecture is built on internal competition and the daily quest to improve everything we do overtime. Professional sportsmen don’t become champions overnight. Instead, they work on the little things that might seem insignificant to outsiders before they reach world-class results. With us, we use our algorithms one by one to analyze data sets and produce forecasts. Then we put them up against each other and let them compete with one another to see which performs best. The next day we do it again, and the day after that again, and again, and again. This helps us to see progress every day, constantly improving our results.
Around half of the methods we use are built upon the work of the great mathematicians of our generation. We use forecasting equations from great minds like Holt, Winters, and Croston. We admire their work, and us being us we have been looking into ways how to improve their equations even more. It has taken us years but we have found our own paths to perfection.
The second half of our methods use principles of AI and mainly deep learning. We use complex 4-layer models that learn datasets and come up with surprising forecasting results. The issue with these methods is that they become effective with only vast amounts of data, and not every product in the customer’s portfolio can produce such quantities.
So for us, it is not about using buzzwords and demanding that only AI is the future. It probably is, but remember, we are data-driven. We measure everything constantly. If AI outperforms standard math, fine! If it doesn’t – all the same. It’s the results that matter, not the marketing headlines or trendy blog posts.
Numbers are just numbers if left alone. They might be fascinating for data geeks like us, but they remain meaningless empty shapes if you can’t read the story behind them. That’s why we have invested so much time into transforming our results into stories with a human face.
The way we create our stories helps us achieve maximum understanding of our data for business owners around the globe. We believe that data science should be understandable to most everyone, and our users shouldn’t need a degree in applied supply chain management to understand our work. The calculations we carry out are complex, but the messages and recommendations behind the numbers should remain simple: buy, sell, stop buying, stop selling.
Ultimately our goal is to make the system as user-friendly as possible, and that applies to our work in data as well as meaningful texts. On the mathematical level, we need to make sure we recommend actions that are not only precise but also relevant to the customer. In the world of words and sentences, we need to find ways for our system to speak and interact in voice command. So much work needs to be done, but there is enough time to perfect our methods. And we are just as excited about the results now as we were at the beginning.