Home | Login | Recent Changes | Search | All Pages | Help

HowToSpeedUpYourProject

We've mentioned several activities for slowing projects down (WhatSlowsYourProjectsDown).

What can we do (other than "Don't do the things to slow the project down") to speed up a project? DonGray 2004.06.09

Do exactly the work (all and only) required for your project. -- JimBullock, 2004.06.09

Clear the track for the project team. Make sure they have:
  • clean, stable, decently performing development and test environments
  • the right tools to do the job and the training to use them
  • a place to meet and places to work quietly and alone
  • plenty of whiteboards
  • enough status reporting for accurate planning and management, but no more than that
  • as few meetings as possible, and only meetings that move the project forward

--FionaCharles 9-Jun-2004

  1. Thou shall identify thy customer(s) and respond to them.
  2. Thou shall have a destination.
  3. Thou shall explore thy customer's requirements
  4. Thou shall create a plan for reaching the destination
  5. Thou shall recruit the appropriate team
  6. Thou shall identify and mitigate risks with the entire team.
  7. Thou shall update the plan to respond to the constraints of the team.
  8. Thou shall make part of the plan the evaluation of progress and correction of direction.
  9. Thou shall follow the plan, evaluate progress, correct direction, and update the plan.
  10. Thou shall continuously consult with the customer(s) about course corrections and plan changes.
  11. Thou shall treat every person with respect.
  12. Thou shall work to bring everyone to the destination safely.
  13. Thou shall have a plan about what thy will do next (after this project).
  14. Thou shall help other teammates create a plan about what they will do next.

SteveSmith 2004.06.09

  • Keep people on only this project.

JohannaRothman 2004.06.09

  • First, find out whether speed is important for this project.

DaleEmery 2004-06-09

  • Don't do things you don't need to be doing. (This may require some small measure of courage.)

DaveSmith 2004-06-10

  • Measure results, not effort.

KeithRay 2004.06.10

  • Cancel it now, rather than later.
  • Extend the ship date now, rather than later.

(Those may be subsets of Dave's advice above.)

MichaelBolton 2004.07.23

I'd like to extend Dale's recommendation:
  • find out what is meant by speed

I suspect the strategies and tactics are different between finishing quickly before the requirements change, and responding quickly to requirement changes.

--DaveLiebreich 2004.07.24

  • Ensure all people involved have meaningful work to do.

Bring them on only when they are needed and let them go when their work is complete.

MikeMelendez 2004.07.27

Building on Dave's comment about speed:

  • find out what is meant by speed.

I suspect the strategies are different when something approximate now is worth more than something precise later, vs. precision having more payoff than delivery time. These several "speeds" amount to fitting the solution to the problem, which begins by understanding what the problem is.

At least as important is figuring out what doesn't matter, usually beyond some threshold, so you can stop talking about that. More subtly, stop trying to optimize for something where all the difference you can possibly make still doesn't matter. Tons of work makes it 5% more right, or 3% faster. Great, but I don't really get a payoff until it's exactly right, or 50% faster, so that's not worth a lot of work, is it?

Which brings the last subtle variation for the moment. If all you can possibly do doesn't make the project "good enough" meaning "fast enough" or "precise enough" or "cool enough" or whatever, you speed up your project by having that conversation right now. Don't delay. "This isn't going to turn out the way we've talked about. What do we want to do?"

-- JimBullock 2004.07.27

Speed kills. The standard practice is to sacrifice all other goals to meet the schedule for finishing the project. If there are no other goals, then just deliver what you have now; that will speed things up tremendously.

Wait. There are other goals? Have you identified all of them? Have you quantified them? Have you made a vague plan which accomplishes all of them to the required levels? Have you specified the tests which will be applied and will compute the numerical result to be compared to the minimum required values for the goals?

I got some of these ideas from Tom Gilb. The small paper is 12 Tough Questions, but the comprehensive 600 page monster is Competitive Engineering which may only be available for PDF download for another day or two, and is freely available in the downloads section of Gilb.com today. Gilb is a smart guy who once got Jerry to do all the work on a book they co-authored after one day of joint discussion.

Dick Karpinski [email protected]

It was the only way it was going to be done, Dick, and for me, a book is not that much work. But thanks for remembering. - JerryWeinberg 2005.06.04
  • Release Criteria or some other technique that allows people to have examples of the minimum project goals. -- JohannaRothman 2005.06.07
  • Don't try to go unnaturally fast. Speed may not always kill, but it causes accidents, and accidents slow down any trip. -- JerryWeinberg 2005.06.07
Appropriate use of "time boxing" can focus a team's attention by subtracting distractions (such as the temptation to plan or design ahead for things that don't fit in the timebox). Inappropriate use of timeboxing can slow things way down by pushing problems under a rug. --DaveSmith 2005.06.28
You can speed up projects by using the smallest number of really good people that can do it in the time you have. More people, working in parallel, can make a project go faster, but the more you have, the more coordination time you have. One person can work very fast, but you may pay later in slowdown of maintenance, or if something happens to that one person. Generally, a proper balance between the benefits of large numbers and the benefits of small numbers is what you need to go as fast as possible, but always with the best people you can lay your hands on. - JerryWeinberg 2005.06.28
  • Notice what's killing focus, and banish the focus killers. Does your cell phone go off during team time? That'll cost you a $1 donation to the after-we-ship party fund. --DaveSmith 2005.07.27
Be careful that the after-we-ship-party fund doesn't become the most profitable part of your business model. But, some other delaying things you could charge for:
  • People dropping into your cube without waiting to be invited.
  • People not responding to key questions on the schedule they promised, so that you have to ask them again.
  • Meetings called with no agenda, or meetings that detour into things that are not on the agenda.
  • What else?

- JerryWeinberg 2005.07.31

Updated: Sunday, July 31, 2005