About

Hi there, welcome to my blog!

I’m Paul, and I like to mess with data.

I’ve always been fond of puzzles and computers. Over time, this developed into a passion for programming and statistics. I particularly enjoy developing systems that leverage, automate or optimize decision-making processes. Thereby preventing the human biases observed by Ariely, Kahneman, Tetlock, and their colleagues. Or allowing businesses to operate just a bit more efficiently and effectively!

Currently, I am managing Impala Solutions, a data consultancy in the Netherlands. Our mission is to help organizations create customer value through data. We do this by building data products tailored to their most pressing business needs. Sometimes, this means we develop and deploy machine learning models that automate operational decisions. Like, which customers should receive an special email offer, or what’s the optimal production level for our factory produce today. At other times, we provide quantitative research and management consultancy to help organizations make important strategic decisions. Like, what customer segment should we focus acquisition on to maximize revenue growth, or how can we change our customer service process to decrease wait times. Nevertheless, we always work towards building data products that produce customer value: be it a strategic report, a cloud dashboard, or a predictive model. Get in touch if you want to learn more!

The past decade, I have been working in the field of data science and analytics at both multinationals and startups. My career started out in the domain of Human Resources where I actually wrote a PhD dissertation on the topic of People Analytics. I strongly believe quasi-experimentation and data science can help organizations improve their strategic policies. For instance, those policies related to the attraction, development, motivation, and retention of people.

During the past few years, I have been developing machine learning models and expert decision systems to optimize business processes. My models have been used to predict claims fraud, set prices dynamically, extract text and data from documents, estimate customer value and churn likelihood, optimize call center workload and capacity, and many other relevant outcomes. If you want to find out how I can help your business move forward, get in touch!

Next to this, I provide professional education in data-driven management, statistical programming, and data visualization. I give masterclasses and workshops at Erasmus University, TIAS, Tilburg University, and the Jheronimus Academy of Data Science and develop e-learnings for the Academy to Innovate HR. I also develop and facilitate learning programs tailored to organizations, like I did for KPN, Capgemini, IntelligenceGroup, Essent, and many Dutch hospitals and governmental agencies. Do reach out if you’d like to know more!

And I host this data science blog of course! I blog mostly about programming, data visualization, and machine learning. Proper data visualization is essential to communicate the enormous heaps of information generated these days. Particularly interactive ones. Where users can freely explore patterns, unhampered by the intentions and biases of the original creators. Regarding machine learning, I am truly amazed at the ever-increasing pace at which computer algorithms are able to both solve global issues, as well as create ethical challenges along the way.