from Hacker News

Ask HN: How does your company plan work?

by afigkka on 12/6/22, 2:27 PM with 1 comments

My original approach to this was work doesn't needed to be planned, it needs to be done, BUT in an org approaching 1000 people who need to be kept busy shuffling papers some amount planning to attempt to coordinate people is needed.

What I would like is project driven teams, who form to solve a problem, deliver a solution, then disband. This doesn't support empire building though.

Current situation at my employer which is starting to transition to a more matrixed organisation:

Ahead of each quarter the Product Development department opens the window for submitting new objectives

These should be tied to the company strategy and are mostly driven by the Product Management function

Each objective will detail the outcomes, teams required and a guess at how many quarters it will take

Then all the teams that are suspected to be involved will consult the objective and estimate how much of their time is needed to deliver the requirements that aren't yet known

Each team then aggregates this in a rough % of their capacity of the projected quarter (like Objective Pickled Giraffe : Team Ostrich 50% Capacity in Q1 required)

Then the submission window closes and all the objectives go to the exec team who pick which ones will be done and which ones need more info to make a decision.

Then we kick off the quarter and start work on the objectives.

This then doesn't fit with any other part of the company who have their own planning cycles and then we fight to pull people in when needed. Hence why we're becoming a 'matrixed org' to solve this.

  • by lucozade on 12/6/22, 4:28 PM

    Your naive approach would work if 1) the possible projects match identically the resource profile without overlap and 2) there aren't any more possible projects than the set that match the resources ie there is no prioritisation decision to be made.

    Unfortunately, that never, ever happens ever. So the organisation needs to have a mechanism for deciding how to prioritise work and to estimate what work can reasonably be done simultaneously given resourcing constraints. And it looks like you've described their approach to those two things.

    The approach you describe sounds reasonable for a medium sized company assuming the quarterly process takes into account on going projects ie they're identifying and scheduling for resource gaps rather than fully re-scheduling every quarter.

    A more common approach with that size or bigger is to use a budgeting approach. So notional dollar values are assigned to projects that are given priority and the project owner funds resource from other teams by giving them budget.