.avif)
The Lightdash Way 🎉
To kick off demo week, I wanted to share our approach to BI at Lightdash and why we think it’s the best way for teams to quickly realise the value in their data and more importantly, make it accessible to everyone in their company.
Making self-serve into a reality
The Lightdash Way is about giving data consumers meaningful building blocks to answer their own data questions. These blocks aren’t tables, joins and columns - they’re revenue, orders, customers and all the entities familiar to our end users. Using these building blocks, our users can focus on answering the question they need, rather than figuring out the right datasets and how to join them. That’s what true data self-serve looks like in an org.
Let’s get into an example:
An e-commerce company receives orders online from its website as well as affiliate partners and its retail presence. To get an overall picture of their total revenue as a company - we need to sum revenue from all the places we sell our product while accounting for incomplete or returned orders.
How do we share that definition of “revenue” with everyone in the company? We have three options:
- The data team could serve all data questions but that doesn’t scale.
- We could give everyone access to the data and let them run SQL/excel/python but errors creep in and the data team jumps in anyway.
- We write the exact definition of revenue as a metric in our semantic layer and let our users query that metric.
The third approach is The Lightdash Way: your data team takes responsibility for maintaining a semantic layer that represents the relevant context to your business. This layer puts our data consumers on-rails by removing the need for them to understand the data logic (in SQL) or the relationships of every underlying table in your data warehouse. By hiding away all the tricky joins and aggregations, we can be confident that our users are always getting the right number when they ask for “revenue”. And then, like magic, that reliable “revenue” metric is available to everyone in Lightdash.
Analysts need better tools for BI
To make this setup work, data teams and Analytics Engineers need the tools to properly own and maintain something as complex as the semantic layer. And once they have them, they give themselves leverage by enabling self-serve for their end users.
We think it’s high time that Analytics Engineers are given better tools. Analytics Engineers tell us they love working from their text editors, they’re savvy on the command line and they work with version control and git all day long.
Lightdash was born from the idea of letting analysts use the tools they they love, and not forcing them to click around in a UI. This introduces our new set of Lightdash developer tools, focused on improving Analytics Engineers’ productivity and shortening the development loop of building your BI and reporting layers.
This is what Demo Week is all about and over the next few days we’ll be announcing:
- Instant developer previews - spin up a new BI project from scratch during dev
- CI/CD workflows - deploy to production when you merge code into main
- Code generation - tools to bootstrap an entire BI suite in seconds
This is our first major step towards giving data professionals the superpowers they need to easily and reliably serve the rest of their organisation.
That's it for today! Come join us in the new Lightdash Community Slack to stay up to date through demo week! 🎉
