HOW do you convince your managers that the solution you have in mind is a good/the best solution?


I need to be able to explain in layman's terms (or maybe, in business terms) WHY the solution is, in fact, a good one. Of course, this makes the implementation of the solution more expensive by default. I have to take more time to research the options, to make sure that everything I say is true (backed by facts).

Looking back on what I just wrote, I don't think it's bad for my manager to be sceptical. This keeps me sharp, and it'll prevent me for going with a cookie cutter solution for the wrong problem, more or less.

What I am looking for is actually the things I need to provide to make a business case. My biggest flaw is that I'm a techie at heart, so my solution is not from a commercial PoV, but from an elegance PoV, if you will. I know I can implement certain solutions, but maybe it's not always needed.

Don't get me wrong. We DO have a backup setup, and we can deal with server outages. We don't have a huge capacity, so we have to be careful on what we spend our hours on. We've got some time every week for random things (free initiative time), so I think I'll take my time there to pursue these issues.

PS: please retag, can't find appropriate tags

  • you may want a soft-skills tag (not enough rep to re tag) Apr 30, 2009 at 13:09
  • You can edit your question and add tags yourself
    – Alex Angas
    May 1, 2009 at 12:28

6 Answers 6


Wrong question. Nothing to do with RAID.

How can you convince your boss that you know what you're doing.

Assuming you're a knowledgeable sysadmin with prior experience, then you should be able to set this up blindfold. If your boss (who does not have to be technical), doesn't recognise the skills you have and bring to the solution then...

You already recognise the risks, and I also assume you can describe these to your boss in a clear and understandable manner.

Remember, you need to get his buy-in, but don't short change yourself. Insist on getting decent (maybe more expensive, branded, supported) kit instead of the cheapest from eBay or PCWorld.

Test. Test. Test.

Then install and relax...

  • 1
    Accepted. I think I will go for setting up a test environment in our 'free initiative' time, to show that we're skilled enough to use and manage it. When we've got that set up, it's basically about getting it covered by SLA fees our clients pay. But honestly, when it's set up it's not more expensive to manage than our current setup, so it's only the setup and testing that costs money. And THAT will be covered by our free time! Apr 30, 2009 at 12:50

Dude, if you can't get traction on this basic stuff, just quit.

edit: if you want to try, try expressing it in dollar terms.

RAID converts a disk failure from a stop-the-world-drop-everything-event to a ill-replace-that-drive-during-the-next-maintenance-window event. It enables you to handle failure more cost effectively.

Server downtime = $$$/hour, hours == the time taken to restore from backup from a failed lun.

  • I think that might work. I think that's my biggest flaw though, my view is always from a technical perspective. I KNOW that it's better, I can explain why, but I have a hard time quantifying it. Thing is that we haven't had any outages for a long time (cross fingers), so it's hard to say 'but what if!'. Maybe I could find some testimonials on the web, see if I can make a case. Just me saying 'but it's better' doesn't really hold. And it shouldn't, really. Apr 30, 2009 at 12:18
  • (As a practical tip) Try to either get hold of a well setup test server or try to get invited to a vendor open day, where you can show your boss how it works. Pull out drives (still running). Insert new drives (still running) etc etc.
    – Guy
    Apr 30, 2009 at 12:46
  • @Erik: No need to say "but what if!" -- it's guaranteed that a hardware failure will happen, it's just a matter of when. If you haven't had one in a long time, then it's even more likely that you will soon. :-)
    – Head Geek
    Apr 30, 2009 at 13:56

You need to cast your arguments in terms your boss can understand. From your description, it sounds like you've got a hard road.

You need to come up with REAL WORLD scenarios you face in your environment and then walk him through what would happen with and without RAID.

I haven't seen it work well when you start talking $$$'s right out of the gate, but you'll need to cover that eventually, so have your figures at the ready.

Good luck, doesn't sound like a good environment.


I think that not every sysadmin is good at persuading people to do things, tinkering with servers and networks is a very different skill set from those needed to persuade people of things.

There is lots of information out there about this, it is soft skill. Some of it is innate, but you can learn techniques to help you be more persuasive. Unfortunately not everybody accepts a well thought out logical argument, in fact sometimes this can hinder you.


When dealing with any non-technical managers who are resistant to changes, your best bet is to speak in terms of $$$$'s. Therefore, create a spreadsheet to contrast any cost outlay of the RAID components to the potential loss of time/customers/sales down the road.

RAID does not replace a data backup solution, so you have to show the cost savings of having more fault tolerant systems. This will usually be in reduction of down time, and simplification of handling disk failures, compared to a full restore from backup.


Express it in terms of time to restoration of services.

If you have good backups, a disk failure won't result in significant data loss, but it will disrupt your operation for hours (if you have spare parts) or days (if you're waiting for a vendor technician). If you have backups + redundancy, you theoretically have no service impact. I say theoretically because as Mr. Atwood's story about RAID with the StackOverflow servers illustrated, RAID isn't always what you think it is.

Ultimately, your boss may not care -- and that is his or her prerogative. Often as a sysadmins we have to live with business decisions that we don't care for.

Not the answer you're looking for? Browse other questions tagged .