I hate chasing down why that dashboard looks off. You hate chasing down why that dashboard looks off. It always starts the same way: a stakeholder pings with a screenshot, something looks broken, and you find yourself deep in someone else’s query trying to track down what changed, when, and why. Meanwhile, trust in your analytics stack quietly erodes.
While the “X as Code” trend has been everywhere for the past few years, it isn’t just a fad. Version control, structured change management, and reproducibility have historically been reserved for data models and pipelines, but not dashboards.
Lightdash provides you the option to truly treat dashboards as code. You can download your dashboards and charts as YAML files, make changes locally, test those changes in preview environments, and push them back into production with a pull request. No more screenshot comparisons or mystery edits.
What does this look like in practice?
Using the Lightdash CLI, you can run a single command to export your entire project’s saved content:
lightdash download
This creates a /lightdash directory in your current folder (often your dbt repo), containing all of your charts and dashboards as individual
.yml files. Want to only edit one dashboard? You can do that too:
lightdash download -d < Instance URL>/< Dashboard UUID>
Want to grab a few specific charts and version them alongside your dbt models? Easy:
lightdash download -c < Instance URL>/< Chart UUID>
Preview Projects
Before pushing any changes live, Lightdash makes it easy to test everything in a Preview Project. These temporary environments are cloned from production allowing you can safely validate changes to your metrics, dimensions, charts, and dashboards.
Creating a preview project is easy if you have developer permissions:
lightdash preview
Preview projects are a core part of the “as code” workflow as they allow you to validate dashboard edits before promoting them to production. Providing a preview project link to your stakeholders can prevent pushing incorrect changes to your production dashboards and will increase trust in your data team.
When you’re ready to deploy, you just upload the changes using the CLI:
lightdash deploy
Why this matters
Lightdash’s approach to dashboards as code solves a huge pain point for analytics teams:
- Accountability: Know exactly who changed what, and when.
- Portability: Spin up staging or demo environments that include production-grade dashboards.
- Reusability: Duplicate dashboards across projects or teams without manual copy-pasting.
- Scalability: Review and approve dashboard changes just like you would with dbt models or transformation logic.
We are excited to see features like these be developed in competing BI platforms.
Final thoughts
This is a better way to work. If you’re spending too much time debugging dashboards, chasing stakeholder feedback, or duplicating work across environments, it’s time to consider a tool that puts dashboard logic where it belongs: in version control.
At Driftwave, we help teams adopt modern analytics tools like Lightdash, so you can spend less time firefighting and more time building. If you’re ready to bring the “as code” philosophy to your dashboards, we’d love to help.