Scrum But (Part 2)

Let's assume that the worst from Part 1 has happened. You've been on a death march. It's not been a pleasant few months trying to curl out a piece of software. How do you keep this from happening again?

  1. Get whoever keeps promising things on a deadline to stop it!

This is going to depend on who is setting the deadline. If the deadline is set in law, you're screwed. If it's contractual, there might be hope...whenever the next contract is written. If someone internally set the deadline, then you have more chance of making some changes. This will depend on the politics at your workplace.

  1. Re-estimate the work as soon as the details are known.

Create and refine (groom has some rather unpleasant connotations, so best not to use that term) all the stories as soon as the details of them become known. It's pretty hard to make a good estimate of work without the details because sometimes changing one detail can drastically increase or decrease the scope. If the vague estimate is 4 months and with all the details that estimate jumps to 10 months, you already have a serious problem. You will need the total points from your new stories, your velocity, and your estimate in order to make your case to management. Excuses may be made in hopes that your velocity will increase over time to cover the gap. The bigger the discrepancy between estimates, the less you should be willing to accept them.

  1. Have the same people who created the estimate do the work.

If Team A creates an estimate for a project, then they will be basing it off their own level of skill and experience with the scope of the project. If the project is then given to Team B, they may not have any knowledge about the background of the project. That estimate will have to be multiplied by 1.5, 2, 3, or even 5, depending on how difficult it is to learn the subject matter. This does force a choice on whether each team should be focused on one part of the system or if each team should have a decent amount of knowledge about the whole system. This gets into the concept of the "bus factor", which could take up pages by itself.

  1. Abandon all hope.

If the option selected from Part 1 is always "death march", then something is very badly wrong. Either your projects are at the bottom of the pile but still have to be done, there's nothing that can be done due to outside constraints, or nobody who can do anything cares enough to change this pattern. Unless this cycle is broken, there will be more death march projects as far as the eye can see. This is not healthy for anyone.

Start looking elsewhere because this will eventually destroy you mentally and then physically. People have had strokes and heart attacks due to burnout related stress. I shit you not. I don't want that to happen to you.

"Remember, I'm pulling for you. We're all in this together." -- Red Green

Day 33 of #100DaysToOffload