“Data science / epidemiology / statistics is 10% inspiration, 90% perspiration, and 20% desperation.”
Why the total is more than 100%? Ever felt this in your work?
The idea is just the tip of the iceberg, while the bulk of the effort—and the occasional panic—happens behind the scenes?
I’m curious to hear from fellow professionals:
How do you balance creativity, hard work, and those moments of “desperation” in your projects?
Any tips for surviving—and thriving—through the messy parts of data and analysis?
Drop your experiences, stories, or even memes in the comments!
Thank you for starting this interesting conversation. It does feel nice when people consider these aspects about individual’s experiences.
For me, the biggest part is that I love what I do. I love being part of DHIS2 and the DHIS2 Community. In addition to that, working with a great team that lifts me up, encourages me, and is always supportive. Furthermore, I’m so lucky that my work involves a really helpful and active global community with members working on their local DHIS2 projects and supporting each other across the global and for different implementations.
I’m always learning, helping, and there’s a balance that I try to include in my life a variety of things that I like and enjoy such as working remotely from a coffee shop or taking a break and try different activities. I try my best to never allow ‘desperation’ knowing that we’re all trying our best.
Hi @Gassim,
Thank you for your thoughtful response—and happy new year
From my experience, one of the biggest challenges is data quality, especially when working under operational pressure. This becomes even more complex when implementations are on older DHIS2 versions, where some functionalities or validations that could help improve data quality aren’t fully available.
What has helped me is accepting that the “messy” part is unavoidable in real-world data work. Taking time to understand data flows, documenting limitations, and leaning on the community for practical workarounds or lessons learned from similar contexts often turns frustration into learning. What hasn’t worked so well is trying to fix everything at once—prioritization really matters.
I think this is where the CoP is particularly valuable: sharing realistic experiences, version-specific solutions, and strategies to improve data quality even when system upgrades aren’t immediately possible.
