We have two different legacy systems (window services in this case) that do exactly the same thing. Both of these systems have small differences for the different applications they serve. Both of these system's core functionality lies within a shared library.

Most of the time, the updates occur in the shared library and we simply deploy the updated library to both of these systems. The systems themselves rarely change.

Since both of these systems do essentially the same thing, our development team would like to consolidate these two systems into a single service.

What can I do to convince management to allocate time for such a task? Some of the points I've noted are:

  • Easier maintenance
  • Decrease testing/QA time

Unfortunately, this isn't enough. They would like us to provide them with hard numbers on the amount of hours this will save in the future and how this will speed up future development. Since most of the work is done in the shared library and the systems themselves never change, it's hard for us to quantify how many hours this will save.

What kind of arguments can I make to justify the extra work to consolidate these systems?

  • "They would like us to provide them with hard numbers". Why are they saying "no" like that? What other, more important work do they want you to focus on? – S.Lott Feb 4 '11 at 20:49
  • The feeling I get is that they don't care about code quality, ease of maintenance, etc.... They want to know how it will assist them in creating "features" to bring in more revenue. If it gets in the way of creating features faster, than it's a road block. – nivlam Feb 4 '11 at 20:57
  • It's clear from the question what they don't care about. My clarification is asking what they do care about. What seems more important than this? To be even more specific, if you said it would take X hours to fix this, what other thing could the invest those X hours in? What is competing for your time? – S.Lott Feb 4 '11 at 21:13
  • Rather than invest time in updating our system, they'd rather invest time into developing more features that they can market. – nivlam Feb 4 '11 at 21:19

How much time do you spend on both of these systems at the moment of the course of a year?

Take that figure and divide by by two.

That should give you a rough estimate of the time you'll save by merging them.

For a more realistic figure use 1.8 or 1.7 as the divisor as there won't really be exactly half the work.

Then take the expected life span of the new system and multiply that in to give you a final figure.

If the time/cost of merging the two systems is less then you've got your argument.

  • Hmm...I can't give a better answer than that +1 – Michael Brown Feb 4 '11 at 20:48

Adding on to Chris's explanation you can also look at the amount of electricity and the cost of cooling required to ensure both of those machines need to stay running. Also the amount of money that is being paid in license renewals and support contracts. If you take those in conjunction with what Chris proposed you should have some broad figure. Don't forget to include inflation in the bottom line.

  • I'd avoid inflation, unless I wanted to get into net present value and the like. – David Thornley Feb 4 '11 at 22:08
  • @David equally true. – Woot4Moo Feb 4 '11 at 22:26

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.