Data & AI
A risk matrix is not a spreadsheet
Why Excel templates break at the first audit, and how to model risk so the number can actually be defended.
Technical notes from real projects: what worked, what did not and why. No smoke, no generic recipes.
Data & AI
Why Excel templates break at the first audit, and how to model risk so the number can actually be defended.
Architecture
Duplicating the report per client does not scale and isolating by workspace gets expensive. Here is how real multi-tenant works, and where the traps are.
Automation
Half the automations we are asked for stop being necessary the moment the process is drawn out. How to tell which ones are actually worth building.
AI agents
Nearly every AI pilot that stalls does so for the same reason, and it has nothing to do with which model was chosen.
Data & AI
Fabric does not replace Power BI or make it obsolete. When it genuinely pays off, and when it means paying for capacity you will not use.
Infrastructure
It is rarely one expensive resource, it is thirty cheap ones nobody turned off. How to find the spend and how to stop it coming back.
Automation
Starting with lists is cheap and often right. The problem is not knowing where the line is, and crossing it without noticing.
Data & AI
Reading tables directly works until the first SAP upgrade. The routes that hold up, and the one with licensing consequences.
Development
Three different takes on agentic development. What each does well, and what to look at before standardising your team on one.
AI agents
The Model Context Protocol has become the standard way to connect agents to systems. What changes, and what to govern before opening it up.
Development
An agent writes in an hour what used to take a day. The bottleneck moved to review, and that is where control gets lost.
If the article sounded familiar, we have probably solved it before.