From Excel to Snowflake: Bridging the Data Culture Gap

(This article is a transcript in english of one of my french linkedin posts)

We must realize that we live in a world where companies still fully operate with Excel to store, analyze, and dashboard data, while on the other side, companies adopt a "modern data stack" with technologies like Snowflake, dbt, Tableau, and even LLM.

When one is immersed daily in a data environment, the Excel-centric approach can seem a bit obsolete. It's a somewhat "manual" approach prone to errors, without real data governance, and above all, difficult to scale.

Of course, it's far from the ideal solution, but if you think about it, it's an approach that can still work today in cases where the volume of available data is still low. In that case, it's mostly a matter of priority, of knowing how to focus on what really creates value for the company.

Because after all, why do we create all this data infrastructure, so complex, if not to create value?

And there too, there is an issue we need to discuss... The value created by these data platforms is not always noticeable by executives and decision makers.

We can't really blame them, because the crux of the problem is that to really perceive this value, you need a certain data culture that is not easy to acquire (the famous data acculturation).

At a minimum, to clarify the situation, one solution would be to design the data platform, from the start, as a product in its own right, define success metrics and track progress. On the other hand, use a business case driven approach to gradually grow the platform. Create value by starting small and then growing progressively.

These elements can help better understand what's going on and keep track.

Then, I also think it's our role, as data practitioners, to make this world more accessible to the "average joe". To show that data provides useful daily insights, that it structures the company, that it provides a competitive edge...

Finally, it may seem trivial, but one of the first barriers to understanding is already data jargon, which is not yet well defined by the way, differs from one company to another and can even cause comprehension problems between different data scientists or data engineers.

The data scientist, data engineer or data strategist roles themselves are still far from being well understood by the majority of people on this planet