[robynsmith]

Source?

I'm a relatively new engineering manager in an older company in Canada (turning itself into fin/insurance tech), and a big part of my role has been about breaking down silos.

The silos where created at a time when being really efficient at 'your piece' was what the system rewarded, whereas now we want to be efficient at shipping code, which means efficient value streams (from inception to production)...which has basically meant many rounds of silo deconstruction.

But I'd be VERY curious about why silos get built in the first place & what incentives are in place to make this happen...and why this is bad. To be honest, maybe I'm just idealistic, but I see silo building as not desirable.

[derivagral]

Not OP, but imo the classic thought is that your managerial influence is directly correlated with the count and quality of employees under you. In this sense, silos get built not quite for specialization, but because your individual managerial incentive is to try to grow you and yours at the expense of (internal) competitors for a fixed budget; interoperation can be a direct threat to your influence and growth.

A famous study of behavior like this might be Kodak, and how internally the film "silo" crushed the digital "silo".

[robynsmith]

Ah interesting.

I think I might just be idealistic and having thought about it this way. Thanks for the context.

[wpietri]

For the background and how things go wrong, you might enjoy the book "Confronting Managerialism" by Spender and Locke. The Lean folks also have a lot of good critiques. A book that sticks in my mind is "Simple Excellence: Organizing and Aligning the Management Team in a Lean Transformation", but there are a bunch. And for software specifically, Mary Poppendieck's work is great, as well as Reinertsen's "Principles of Product Development Flow".

[robynsmith]

Thanks for the recommendations!