0

I am getting ready to head up an internal project. I already know that the budget will be tight on this. I am looking for effective ways to reduce project cost. What I would like to see is the most code for the least amount of budget.

For example, if I develop an example page and have an Intern or Jr. Developer refactor another page in the same style the hours on budget would be less assuming they code at the same rate as me. However, I know that Interns and Jr. Developers will not code at the same rate as me. Giving up high cost functionality seems like another way to go, but in the end all functionality is desired. As of now, there is not a time deadline, but that can at a moments notice.

To try to avoid this question being closed, I would like to see answers with the following...

  1. What has worked for you in the past.
  2. Why a particular process worked for you.
  3. What has not worked, i.e. you tried something you thought would work and you were surprised it did not.
| improve this question | |
  • I am interested in how you would discriminate reported cost reduction actions and a placebo effect or attribution of a random result. – David Espina Jan 22 '14 at 19:49
  • @DavidEspina Here is the situation. Requests are coming in for improvements. The person making the requests is also saying stay under this dollar amount, but still give everything that I ask for. I am looking for the best possible way to give them everything they are asking for and staying in the budget. So in the end the reported cost should be all that matters. I just do not want to go down a path that ends up costing me more because I thought it would save me budget. – JabberwockyDecompiler Jan 22 '14 at 20:22
  • If you are being asked to create more features without increasing the time & cost to do it, you are being asked the impossible. So say "no". – David Arno Dec 22 '14 at 13:53
1

On question 1

  • Reduce scope.
  • Get backing of a CxO.
  • Get rid of team members unproductive and not needed to maintain the software afterwards (or ask the people manager a refund for their hours).
  • Get a champion for each discipline .
  • Deliver early whole processing chain to avoid surprises and risks forgotten.
  • Use prioritizing approach such as MoSCoW.
  • Make it fun.

On question 2

  • Reduce scope; there is nothing like it for reducing costs when everything else has been fixed such as duration, tools, employees, testers, etc.

On question 3 - Get more project team members does not generally help.

| improve this answer | |
  • Thank you for your feedback, I can see that you have run across this before. MoSCoW, looks perfect, I was only sorting in must have and nice to have, I will give that a try. However, in my case, we are a smaller team and I will be the champion and the CxO is already involved (the one making the requests and saying stay under the budget). With that in mind it sounds like I just need to push harder for a ranking of what is most desired from the CxO. – JabberwockyDecompiler Jan 22 '14 at 21:07
  • Good to hear I could help you. With a small team and an CxO involved, that is the best you can get. A good meal (I often go for a drink or food or a walk) helps thinking about good ways to improve performance and cut budget. Just like solving a programming problem can sometimes be solved by just doing something really silly for 10 minutes. Good luck on the project. – Guido Leenders Jan 22 '14 at 21:40
  • Does this count as silly? – JabberwockyDecompiler Jan 23 '14 at 14:56
  • Definitely! Or one of the videos of Dave Platter on software engineering – Guido Leenders Jan 23 '14 at 16:14
1

Quality, time, cost. Pick two.

Even when you pick two, performance of each variable is still very probabilistic. And we have less control over this performance than we like to admit.

If you are looking at a single project, then you are not really looking for significant cost take out ideas or capability game changing initiatives. If you have this type of pressure of delivering everything the customer wants at a fixed, unmovable cost and likely a fixed schedule, what you really need to do is monitor your work extremely closely, looking for the first signs of unfavorable performance variances, mitigating where you can early and often, and then communicating where you are with your stakeholders early and often. It becomes more about communicating early that you are overrunning than actually hitting your promised targets, so that your customers can make the decisions they need to make if cost is such a huge factor for them.

Use proven cost control and risk management methods. Look for tangible evidence, not just what your team tells you.

| improve this answer | |
  • This is very helpful advice. Thank you for taking the time to put it together. – JabberwockyDecompiler Jan 23 '14 at 14:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.