I’m available for freelance jobs. Please visit my Upwork account or use the contact form.

Latest Posts

Tidyverse Presentation

The slides of my presentation at Köln R User Meetup on 14 Oct 2016. It covers an introduction to tidy...

Circle Plots with GGPLOT2

(Github repo) Pretty circular plots. I’m not sure there is an added benefit for the data analyst herself but if...

Loading Shiny Apps Fast with Feather

Shiny Feather When I think about feathers (the real world thing) I associate with the words ‘light’ and ‘slow’. The...

Building R Packages with Devtools

Notes Github Repo In this document (*) means that there is a full example of usage in R in the...

Data Science

I believe in strategically applied Data Science that is implemented not as a (Big Data) Hype but rather as a tool to move an organisation forward in achieving it’s mission.

Based on Bernard Marr’s SMART Data I cover 5 topics with tools and checklists to ensure Data Science is applied effectively and efficiently.

Strategy, Data, Analytics, Visualisation (Reporting) and Monitoring Implementation.

I work solely in R, the home of the tidyverse! Systematically sticking to the tidy data principles results in effective, easy to read R code. The combination of the dplyr, tidyr, purrr packages in combination with personal custom functions are the perfect preparation for any of your desires. Whether your desires are heavy machine learning, static reports or interactive visuals. The tidyverse does it all with its numerous fitting packages.

I also build Data Products in R. Let me take of building your packages, functional coding, Shiny apps and R Markdown.


Start with a question that you would like answered. Use Bernard Marr’s Strategy Board Tool to support you to ask the right questions for your organisation. Once the goal has been decided there are a few ethics/standards (with tools) that span across data, analytics and reporting that are good to adhere to.

Strategy Tools

  • Strategy Board
  • Reproducible research
  • Tidyverse
  • Github Flow


Identify your data requirements, locate and gather the data, import the data and tidy it. The two main principles: First, garbage in garbage out. If the data is no good the analysis won’t fix it. Second, stick to the tidyverse. The importance of consistently preparing your data in a structured style is easily under-estimated.


  • Data sheet
  • Tidyverse (import, clean and tidy data)


The fun and creative part, right? A solid analysis should follow decision making tools that allow it to effectively bring insights that your (strategic) questions needed to know.


  • Exploratory visualisation
  • Regression
  • Machine Learning models
  • Machine Learning decision making tools
  • Managing Multiple Models (Tidyverse)


Answer the question. Let me repeat that more specifically, answer the question such that the answer is obvious for the person who needs to know. That’s all. The task of the report doesn’t change whether you create static reports, infographics, interactive visualisations or Shiny web applications.


  • R Markdown (pdf, html, Word)
  • Shiny Applications
  • Checklist for successful visualisations