Dataharvest 2026 - the European Investigative Journalism Conference

In 2014, we decided to request a comprehensive dataset from the Swiss state: We asked for a detailed overview of all granted export licenses of arms and military goods. Over ten years later, we run the most comprehensive arms exports database in a European country based on the documents received: www.rüstungsreport.ch. In our workshop, we take participants on a journey of investigative research, legal struggles, methods, and outcomes.

We'll explain how we managed to create a database of all arms export licenses obtained for armaments and surveillance equipment from Swiss-based companies around the world. We'll take you through a lengthy access process via freedom of information requests. The information was obtained only after a long legal battle against the Swiss state, which we won before the highest Swiss court. We'll also show how we built a database, organised the data, and how this became a tool of transparency, as we kept requesting the data and updating the database every year. Finally, we will showcase stories that came out of the database, uncovering critical company deals and state practices, and their potential for cross-border stories. We'll guide the participants through our investigations into the use of PC-12 from the company Pilatus in the US war in Afghanistan. And how the toothless Swiss export controls in the dual-use sector, coupled with the strong lobbying of the arms industry, made it possible for the Russian war machine to rely on Swiss high-tech.

This session will enable the participants to adapt our methodology to their countries of interest and give them ideas for further investigations into the defence industry. We will hand out a reader with the most important learnings and a template of the freedom of information request we made.

Coding has long been a skill journalists wanted to learn to make their investigations more efficient and rigorous. The main barrier was the significant time investment required to develop that skill. But since large language models emerged, we no longer need to write code ourselves. We do, however, still need to make informed choices when instructing an LLM to write code for us. Otherwise, those choices get made for us by the model.

How do we instruct the LLM best? How can we understand a code? And how do we catch potential mistakes? No prior coding knowledge is required to attend this session. You'll learn a simple, systematic approach to conversations, context management, and effective prompting that will help you to code anything. The participants should have an account with a large language model provider (ChatGPT, Claude, Gemini or similar).

Journalists often work with sensitive information. This information should not end up in web-based tools like ChatGPT and similar services. However, there are alternatives: local LLMs that run on your own computer. This not only ensures data protection when processing large volumes of documents, but it can also save costs on expensive APIs.

This introductory workshop aims to answer the most important questions: What hardware do I need? What frameworks are available (LM Studio, Ollama, etc.)? Which models can I use for which tasks? And what does such a workflow look like (e.g., with Python)? This session is a mix of presentation and hands-on elements.

To attend this session, no prior knowledge is required. If you want to participate in the hands-on parts, make sure to download and install Ollama and/or LM Studio and download a local model like Qwen3.5-4B

After attending this session, the participants will understand the pros and cons of using local AI models and get ideas from real-life examples on how to use this knowledge.

You need to harvest data from a Web site. But there's no download button. It's time to scrape! There are many options, but one of the most consistently effective is launching an automated browser. You tell the browser where to go and what to click, and when to ingest the content. To follow along, participants should have some knowledge of coding in any language.

Participants will come away from this class knowing the basics of Web scraping with a browser emulator. To follow along, participants should have R Studio installed https://posit.co/download/rstudio-desktop/, create a new project, download this file selenium-server-standalone-3.5.3.jar into the project directory, and have the appropriate Chrome binary downloaded into the directory https://googlechromelabs.github.io/chrome-for-testing/last-known-good-versions-with-downloads.json

Web scraping is a powerful way to access otherwise unavailable data, but it’s becoming more complex as websites deploy defenses like Captchas and anti-bot systems. At SWR Data Lab, we’ve tackled this across investigations ranging from Google price comparisons to healthcare platforms and social media scraping, each requiring a different approach. In this session, we share a practical decision framework for choosing the right scraping strategy based on robustness, cost, and maintainability.

In this session, we will present a decision framework for selecting the right scraping strategy based on our learnings. Rather than promoting a single tool, we want to focus on choosing the right approach for your use case, considering robustness, cost, and maintainability in a newsroom context. Using real examples, we walk through our workflow: from analyzing sites with dev tools to selecting between HTTP scraping, browser automation, and advanced tools—along with best practices and when paid services are worth it.

To follow along, you should have some experience in scraping and, ideally, Python. The participants will be able to extend their toolkit, make smarter choices in their scraping workflow, and handle real-world obstacles efficiently. No special tools are required to follow along

Have you ever tried to investigate how much groceries or rent in your city really impact people’s budgets? Journalists don’t always get all the data in one place. Often, we find it in ads, public announcements, or different sources, then clean, structure, and track it over time, compare it with other datasets, or monitor changes to uncover trends.

This hands-on workshop teaches journalists how to clean, transform, and structure real newsroom data using Python. Participants will learn practical techniques to handle messy data, including changing data types, filtering by values or dates, splitting columns, labeling and recoding, calculating averages and percentages, and extracting quantities from text fields. The session also covers tasks specific to regional datasets, such as converting scripts from Cyrillic to Latin.

With these skills, journalists can analyze grocery prices and compare them with income data or calculate meal costs to report on rising food prices, examine traffic accident data near schools, or track public officials’ gifts and benefits. By the end of the workshop, participants will have concrete tools and workflows to turn raw data into reliable sources ready for investigation and reporting.

To follow along, participants should have some experience with Python basics and working with datasets. After attending this session, participants will be able to turn messy data into clean, reliable sources, compare thousands of entries, and extract insights for investigative stories using Python. Participants should have Python installed on their own computers to follow along. This tutorial can also be accessed via Google Colab, where most of the steps are similar, though Python installed locally is the recommended option for a smoother experience.

There's a (genuine) story hiding in plain sight. Using a mixture of OSINT skills and clues hidden in the room, you will have 30 minutes to get to the story. First come, first served, max 3-4 teams per session competing to get the exit code and get out!

There's a (genuine) story hiding in plain sight. Using a mixture of OSINT skills and clues hidden in the room, you will have 30 minutes to get to the story. First come, first served, max 3-4 teams per session competing to get the exit code and get out!

Want to see change from space and turn it into an interactive story? In this hands-on session, participants will use Copernicus Browser to locate areas of interest, upload their borders, and export high-resolution images. These images will then be turned into interactive sliders in Flourish to clearly show environmental changes, infrastructure expansion, or land-use transformations over time.

Using a real investigative example from a mountain area in Greece that was untouched in 2022 and became an active wind farm construction zone by 2025, participants will see how this workflow avoids common challenges, simplifies the process, and produces accurate visual comparisons.

The session also shows how before-and-after sliders can guide field reporting and support verification, helping journalists confirm that what is observed on the ground matches remote analysis. By the end of the session, participants will have learned a practical, reproducible workflow to turn satellite imagery into accurate visual comparisons that strengthen investigative storytelling and field verification, without requiring complex satellite analysis skills.

Prerequisites:

- To attend this session, no prior knowledge is required. Familiarity with QGIS, Copernicus Browser, or Flourish is helpful but not necessary.

- Tools: Participants should create a free Copernicus Browser account: https://www.copernicus.eu/en

- Participants should create a free Flourish account: https://app.flourish.studio/login

Optional: install QGIS for further exploration of satellite imagery (not required for the workflow demonstrated): https://qgis.org/download/

According to the OECD, 79% of companies use algorithmic management tools to automate recruitment, manage human resources and, above all, monitor and control workers. Subjected to constant algorithmic surveillance, salespeople in all sectors are always aware that their every move is being scrutinized. In digital professions, particularly video game
design, this digital activity tracking fuels burnout and workplace suffering. Delivery workers, especially those on bicycles, face even more serious consequences: the accelerated pace of their work leads to serious workplace accidents and even death. In France, the press has documented more than twenty fatal accidents since the arrival of Uber 
Eats and Deliveroo to the market in 2015.

How are these technologies deployed? How can we bypass corporate communications to document their real impact? How can we identify the hundreds of small businesses worldwide that market these devices? In this session, Clément Pouré, who has published around fifty investigations on workplace surveillance in the Age of AI, and a book on the same topic, will shed light on the surveillance practices of numerous multinationals such as McDonald's and TP (a world leader in call centers), but also on the more insidious practices of smaller companies across all sectors.

In this session, we'll briefly look into the history of workplace surveillance and its impact on employees, and chart an overview of the current technologies and the risks they represent. We will review the methods and sources that can be used to investigate the subject. The session is suitable for beginners, and no prior experience in algorithmic investigations is required to attend.