Martyn Rhisiart Jones

Madrid, Saturday 7th February 2026

What to say?

I was reading an article. It was written by Jeff Wilts and recommended by Bill Inmon. I got to this statement: “Teradata is a full-featured enterprise data warehouse.”  For me, it went further downhill from there.

But this was the coup de grace: “Databricks is a unified data platform that can behave like a data warehouse.” 

I hope seasoned data warehousing professionals get what I am alluding to; if not, here are some more clues.

“Once we migrate to Snowflake, the business will finally trust the data.”
– Senior data architect, global retailer

Business users do not mistrust data because it is on-premise. They mistrust it because revenue does not match finance. Customer counts shift between meetings. “Active” means one thing in marketing and another in operations. No amount of elastic compute resolves a semantic argument.

“We’ve fully normalised the enterprise data model.”
– Data modelling lead, financial services firm

In many organisations, the data model is immaculate and exhaustive. However, it is largely unused. It is consulted mainly by other data professionals. It is like a legal code written in a dead language.

“We can’t release that dataset yet; it hasn’t been fully governed.”
– Head of data governance, healthcare group

A warehouse that is perfectly governed and practically ignored is not a triumph of stewardship. It is an expensive museum.

“The data warehouse is the single source of truth.”
– CIO, manufacturing conglomerate

Declaring a single source of truth without first negotiating meaning is like declaring peace without ending the war. The dashboard loads quickly; the meeting does not end sooner.

“We’re ingesting over 40 terabytes a day.”
– Data platform lead, media company

A warehouse that answers ten critical questions reliably is more valuable. It’s more valuable than one that stores every click ever made, beautifully partitioned and rarely consulted.

Yet status in the profession still accrues to those who build cathedrals, not chapels.

The Persistent Category Error

At heart, these confusions stem from a single mistake. Data warehousing is being treated as an IT system. It should be viewed as a business sense-making apparatus.

Technology can be bought, abused and misused. Understanding must be negotiated. The former is faster, cleaner, and far more comfortable for professionals trained to build systems rather than broker meaning.

Until data warehousing professionals are rewarded less for architectural purity and more for commercial clarity, the pattern will persist. The shelves will grow ever more sophisticated. The stock will remain poorly understood.

And the business, polite yet unconvinced, will keep asking the same awkward question time and time again:

“Yes, but which number do I trust?”

Tools, products, and technology are tangible. Data, information, knowledge, and decision-making aren’t generally easy to grasp. These concepts tend to get confused.

Many thanks for reading.

PS When I am looking for reliable technology Databricks is not on my list.

_😺  Click for the last 100 Good Strat articles… _😺