5 Comments
User's avatar
Alex S's avatar

A couple of observations on this..

First is around completeness of the model. It feels like it’s missing a viable vector: alignment (A), which some may call politics but P is already taken :). This is the person who works across the organization, building relationships, securing support, and making “deals” with the people who matter. In an ideal version, they’re doing this in service of the organization rather than themselves. It overlaps somewhat with the original P, but feels distinct from building and growing a team internally. This is more about external alignment. It also overlaps somewhat with strategy and operations, but again seems distinct enough from any of these. And as you go up levels, you tend to see more and more people who fit this pattern.

Second is that the model treats the axes symmetrically, but P seems structurally different. It’s the only one that can compensate directly for weaknesses in the others. If you’re strong in hiring and growing people, you can bring in O, S, T, and even A. It’s much harder to see how strength in O, S, or T can make up for gaps in the other areas without relying on existing organizational support or someone nearby who is effectively covering those functions.

Philip Su's avatar

Insightful. Thanks for sharing!

Om Prakash Pant's avatar

The Operations dimension is the most underrated one in AI delivery too. The retail AI PoCs that stall after the demo almost always had strong strategy and credible technology.

What they lacked was someone who thought in systems - who asked not just whether the agent worked but whether the organisation could operate it reliably at 2am on a Saturday when something breaks.

Building the machine that builds the machine is the part nobody budgets for because it does not show up in the demo.

timojhen's avatar

Great summary of a model many learn over years. Kudos

8Lee's avatar

Not a new concept to me here but I love the reminder that technology is the last part because so many folks — especially non-technical folks — jump to implementation first and then try to reverse engineer their mission and vision backwards, up the stack.

But, we know that never works. Or, at least those who are informed.