The part that resonates is the displacement of complexity. In most orgs, that complexity doesn’t disappear - it migrates into governance, monitoring, and exception handling.
AI-first often looks efficient at the executive layer because failure costs aren’t visible yet. But once probabilistic systems hit real workflows, the oversight surface expands slowly.
The org doesn’t get smaller. It gets differently structured.
Thank you for such a balanced take on the Salesforce situation. I absolutely think that a data-first approach is sustainable and effective. I think the problem with Salesforce is that they got rid of a lot of their institutional knowledge, which would be very difficult for them to buy back.
Thank you so much! As a data scientist, I’d add that the loss of institutional knowledge doesn’t just slow teams down, it also affects how effectively data can be used. Models and dashboards rely on context, definitions, and historical nuance that rarely live in documentation. When that context disappears, even strong data capabilities take longer to mature, because teams have to rebuild understanding before they can fully leverage a data-first strategy.
The part that resonates is the displacement of complexity. In most orgs, that complexity doesn’t disappear - it migrates into governance, monitoring, and exception handling.
AI-first often looks efficient at the executive layer because failure costs aren’t visible yet. But once probabilistic systems hit real workflows, the oversight surface expands slowly.
The org doesn’t get smaller. It gets differently structured.
Thank you for such a balanced take on the Salesforce situation. I absolutely think that a data-first approach is sustainable and effective. I think the problem with Salesforce is that they got rid of a lot of their institutional knowledge, which would be very difficult for them to buy back.
Thank you so much! As a data scientist, I’d add that the loss of institutional knowledge doesn’t just slow teams down, it also affects how effectively data can be used. Models and dashboards rely on context, definitions, and historical nuance that rarely live in documentation. When that context disappears, even strong data capabilities take longer to mature, because teams have to rebuild understanding before they can fully leverage a data-first strategy.