From fragmented spreadsheets to actionable insights—how NRC leverages CKAN’s open-source power for better decision-making in crisis response.
The struggle is not just against crisis. It is against indifference.
You cannot respond to what you cannot see.
You cannot fight for justice if the facts are buried.
And in a world where data decides who gets help and who is forgotten, access to that data isn’t just about efficiency—it’s an act of power.
Humanitarian data is supposed to illuminate suffering, to make it legible, actionable, impossible to ignore. More manageable.
It should cut through uncertainty, guiding decisions with clarity. But more often than not, it does the opposite.
It vanishes.
Buried in outdated spreadsheets. Scattered across disconnected reports. Hoarded in systems designed to store but not to share.
Like smoke in a broken machine—there, but inaccessible, unusable, ungraspable.
Take a refugee camp in Chad. A humanitarian worker walks through the camp, phone in hand, logging critical data. Somewhere else, another team member does the same—except they’re using pen and paper, scribbling notes they’ll transcribe later.
Multiply this across 40+ countries.
You don’t get "big data." You get big inconsistency.
The Norwegian Refugee Council (NRC) faced this paradox. They had data, but it was nowhere.
Fragmented across spreadsheets, locked in forgotten reports, tangled in systems that assumed stability in a world defined by instability.
They didn’t need another archive of numbers.
They needed a system that made data move—that let people find it, use it, act on it.
That’s why they turned to CKAN.
Not to “store” data, but to make it usable.
To remove the friction, the gatekeeping. And to cut through the opacity that so often turns information into an obstacle instead of a tool.
A few weeks ago, I sat down (virtually 😉) with Nadine Levin, Data Product Manager at the Norwegian Refugee Council (NRC) to talk about what it means to build a system where data informs action, not just archives history. She previously shared NRC’s journey during CKAN Monthly Live—you can watch her presentation here: Norwegian Refugee Council: CKAN in a Humanitarian Context. Or, read our recap blog post: How NRC is Using CKAN to Improve Humanitarian Data Management.
“Why CKAN?” I asked Nadine
She didn’t start with infrastructure. She didn’t mention buzzwords.
Instead, she gave me a philosophy:
“We chose CKAN primarily because of the open-source community,” she said.
Not the software. The people.
A global network of developers, organizations, and users meant that NRC didn’t have to build everything from scratch. They could stand on the shoulders of others.
“We could build a technical product fast by using a “shared brain” meaning we could leverage collective expertise, best practices, and existing solutions from the community,” Nadine explained.
That meant faster development, fewer mistakes, and the ability to customize what actually needed customization.
Still, humanitarian data isn’t like corporate data.
It comes with context, history, urgency. It isn’t just numbers—it’s stories, crises, responses.
So CKAN’s ability to customize metadata schemas became key:
"The ability to heavily customize the information architecture and metadata schema allowed us to add the contextual information that is greatly needed in complex humanitarian data and ultimately made data centralization easier," Nadine explained.
Organizations often assume they need heavy customization to make a data catalog work. NRC took the opposite approach—start with CKAN’s out-of-the-box features and only build what’s truly necessary.
I asked Nadine:
She was clear: less than expected. "With help from OKF, we were able to get a vanilla instance of CKAN up and running ASAP so we could experiment with what features are already included."
CKAN already has so much!" – Nadine Levin, NRC
Rather than jumping straight into custom development, NRC tested CKAN’s core capabilities first. After interviewing internal users and assessing what was already functional, they realized only about 20% of their system needed customization.
That small percentage focused on enhancing usability and accessibility. Two CKAN extensions proved especially useful:
✅ ckanext-announcements – Allowed NRC to communicate updates directly within the CKAN front end, ensuring users stayed informed without external emails.
✅ ckanext-pdfview – Made document-based datasets more accessible by enabling inline previews, reducing unnecessary downloads.
This approach—leaning on CKAN’s strengths instead of overbuilding—saved time, reduced complexity, and made maintenance easier in the long run.
I asked Nadine for a real-world example of how CKAN changed the way NRC operates. Did it truly make a difference in decision-making?
Her answer wasn’t about just storing data better—it was about freeing people from manual work and making information usable when it matters most.
"Our CKAN instance is part of a broader effort to create a data-driven culture at NRC. One key initiative was to create a PostgreSQL connection to automate dataset updates into CKAN through the rich API." – Nadine Levin, NRC
Before CKAN, a crucial dataset—Country Office Fact Sheets—was scattered. It contained historical data on NRC's activities across different countries: number of staff, types of projects, key indicators. But accessing it was a painfully slow, manual process.
"Previously, this data was held in various locations and required days of manual Excel work."
Imagine that—critical information, needed for high-stakes leadership decisions, buried in disconnected files that had to be manually compiled every time someone needed a report.
Now?
✅ Fully automated updates using PostgreSQL, eliminating delays and human error.
✅ Data is visualized directly inside CKAN, giving leadership instant insights instead of spreadsheets.
✅ Senior teams use it for country visits and strategic planning—instead of waiting for outdated reports.
"Now it’s automated, visualized, and directly used by senior leadership for country visits and strategic planning through CKAN!" – Nadine Levin, NRC
Not everything in a tech project goes exactly as planned. Some features turn out to be far more valuable than expected—the kind of improvements you didn’t originally envision but can’t imagine working without once you have them.
I asked Nadine:
Her answer was immediate: Power BI integration.
"We custom-developed the ability to embed Power BI dashboard iFrames at the dataset level." – Nadine Levin, NRC
NRC already relied heavily on Power BI for analytics and reporting. But before CKAN, using these dashboards often meant switching between tools, downloading files, or manually matching datasets. It was inefficient.
So, NRC embeded Power BI dashboards directly into CKAN.
"Since NRC heavily relies on Power BI, this feature has been very powerful for us. It enables colleagues to interact with data dynamically and gain insights without reviewing raw files."
Now, instead of downloading reports or navigating through multiple platforms, NRC teams see live, interactive Power BI dashboards inside CKAN—right where they already manage their data.
Some datasets don’t even have separate files attached anymore—just an embedded Power BI dashboard.
"Some of our datasets are even simply embedded Power BI dashboards without a resource attached because it allows us to still leverage the powerful CKAN metadata schema."
This unexpected feature bridged the gap between static datasets and live analytics, making CKAN not just a data catalog but a true workspace for decision-making.
What started as a simple catalog became a platform where people don’t just find data—they use it.
Every tool has its limits. Even the most powerful platforms leave users wishing for that one missing feature. So I asked Nadine:
Her answer? Built-in data visualizations.
"We are already in the process of developing an in-house visualization tool using a CKAN dataset as the backend data source. Custom visualizations would be a great feature for CKAN!" – Nadine Levin, NRC
Right now, NRC uses CKAN as their “source of truth” for finalized, validated datasets. But once the data is cleaned and structured, many teams need quick, reliable ways to generate visuals without exporting files or switching platforms.
"Since we use CKAN for our curated and finalized ‘source of truth’ datasets, we would like to create visualizations on this clean data."
For NRC, visualizations aren’t just a nice-to-have—they’re critical for storytelling. Donor reports, internal planning, field operations—all of these require clear, compelling visuals to turn raw numbers into actionable insights.
"We use data visualizations in many aspects of our work, like donor reporting, so having an easy-to-use data visualization builder at the resource level would allow non-technical colleagues to have a ready-made visual using the validated CKAN data."
For now, NRC is building their own solution. But Nadine’s wish is clear: A CKAN-native visualization tool would take the platform to the next level.
Technology projects often fail not because the tools are wrong, but because the approach is. Some organizations overcomplicate things, sinking time and money into unnecessary features. Others underestimate their needs, leaving teams stuck with systems that don’t actually solve the problem.
So I asked Nadine:
Her answer was simple, but essential: Keep it lean.
"Don’t overcomplicate your implementation. CKAN’s vanilla features often solve 80-90% of your needs." – Nadine Levin, NRC
Many organizations assume that they need a heavily customized CKAN setup to meet their unique requirements. But NRC took a different approach. They started with CKAN’s out-of-the-box features and only built what was absolutely necessary.
"Custom development can be costly and challenging to maintain. CKAN and the community may have already solved the problem you are trying to tackle. Try reaching out to others and ask!"
This is a mindset shift. Instead of reinventing the wheel, organizations should tap into the CKAN community, where solutions often already exist. The open-source ecosystem is filled with extensions, shared knowledge, and people who have solved similar challenges before.
The key takeaway? Start simple, iterate fast, and don’t build what you don’t need.
Discover how the Norwegian Refugee Council (NRC) is using CKAN to streamline humanitarian data management, improve collaboration, and enhance data accessibility for field teams across 40+ countries.
CKAN sites can now easily expose their datasets to ML pipelines with the help of the newly released croissant plugin.