3 Comments
User's avatar
Madison Mae's avatar

I really enjoyed this read! I totally see the value of the semantic layer, but am still skeptical. Most companies don’t have a solid foundation of data models. In order to have a solid semantic layer, you need the underlying data models to power it. I’m most worried about companies spending all their time investing in a semantic layer when their data models underlying them are complete garbage. Then you’re creating a whole other layer of tech debt that someone will need to fix in the future…. Lots to unpack here for sure!

Expand full comment
Andres Vourakis's avatar

That's a good point! That's why we need data/analytics engineers to keep us grounded.

My company has invested the past 8 months building an information model and rebuilding our data warehouse, so for us, introducing a semantic layer is like the cherry on top. But I agree, a semantic layer doesn’t work unless the underlying data is clean, modeled with intention, and actually trusted by the people who use it.

Would love to have a deeper discussion with you on this topic. I'm sure my audience would benefit from your perspective.

Expand full comment
Madison Mae's avatar

That would be great! Let’s do it

Expand full comment