[lifeisstillgood]

Overwhelmed Project Managers to me implies poor processes - but we all seem to accept it.

Frankly there should be no need for a PM to do anything except in face of disaster -

- plan the work - do the work - realise the pace of work won't hit the milestones at the dates predicted - ok - adjust

None of this needs a PM - just a few hours of a team lead. (ok so maybe that is a PM). but basically if you have one ticket system, and one set of milestones, you can see everything you need a Proiect manager for (need a ops manager or a dev lead or hiring lead - yes but no not project management)

just don't do deadline driven projects and pretty much everything else falls into place

[gibolt]

Product managers shouldn't just be watching a product team execute, but planning integrations, negotiating buy in from other teams, and planning how future features/integrations can proceed.

Building a product is a human process.

[ignoramous]

> just don't do deadline driven projects and pretty much everything else falls into place

Basecamp's Shape Up [0] talks abt fixed 6 week cycles for projects. Trim down projects to fit 6 weeks without compromise and not the other way around. Alternate engineers between heavy product development (6 weeks) and cool-down maintenance cycles (2 weeks).

A good read.

[0] https://news.ycombinator.com/item?id=20408514

[ska]

  Frankly there should be no need for a PM to do anything except in face of disaster -

I think this severely underestimates the importance of a PM role. Project management is like many other things, even if you don't think about it and put time into it you are doing it - just probably ineffectively.

[k__]

Most project managers I met were just the punching bags between customer and devs/designers.

The good one absorbed it, the bad ones kicked down.

[rodolphoarruda]

Yes, they are the first sign that the company oversells its products/services. As soon as the customer realizes his expectations are not being met -- or never will be -- he starts punching the PM. Mature organizations would put Sales in project status calls to speak with the customer in the first sign of misaligned expectations, and that helps a bit.

[rodolphoarruda]

>just don't do deadline driven projects and pretty much everything else falls into place

I've been doing PM work since 2002. I don't think you can work without deadlines -- plural -- in any kind of project for an established business these days. If you think you can, chances are the project is not that relevant/strategic/critical to the business after all. "It can wait".

[lifeisstillgood]

It takes the whole business to stop being deadline driven - and that is hard where exceptions are based on deadlines and upper management manages on exceptions

Basically we need a project system that tells you, based on velocity and expected effort to complete (ie what you get from scrums) when your work will be complete - and if you don't like it you start to adjust.

[giancarlostoro]

One thing that strikes me as foolish:

Giving an estimate without getting all cards on the table. If you have never touched the tech before, and it's new to all your engineers, and you don't invest time into figuring out effort and limitations, you will miss your estimate entirely.

Having an experienced developer evaluate parts of the projects is a good way to gauge the feasibility ahead of time. I went from one job that did this, to one that didn't, I'm back at the former as a result. You can't estimate work on things you've never done with technology that was birthed just yesterday, or lacks an open / public community.

To those PM's that estimate blindly I just want to say:

Good luck, we're all counting on you.[0]

[0]: https://www.youtube.com/watch?v=SmHeP9Sve48

[notJim]

I took PM to mean product manager, not project manager.