Tell me more ×
Programmers Stack Exchange is a question and answer site for professional programmers interested in conceptual questions about software development. It's 100% free, no registration required.
up vote 35 down vote favorite
11

I’m working on a project and my computer is quite slow. I would work more efficient and get less irritated if I get a better developer machine.

How can I convince my boss to give me a better developer machine?

share|improve this question
7  
It's kind of an aside but if he's not responsive you might want to consider the fact that he's not being difficult, he may genuinely not have the ability to fulfil the request. Many companies limit managers as much as they limit programmers. – Jon Hopkins Dec 16 '10 at 10:51
@Jon Hopkins I agree! I can add to that that many companies have different budgets for equipment’s. So even if you are consultant and cost 1000$/day they still can’t give you a better equipment since the budgets differ. – Amir Rezaei Dec 16 '10 at 18:46
Is your boss also a programmer? IT guy? General non-technical manager? My boss is a systems engineer, and he gave me a pretty bad machine. So I re-installed everything, and now it runs fine. He was happy to see me put in the effort to lessen the budget needs, because he just had it cut in half. – Pablo Dec 17 '10 at 2:52

9 Answers

up vote 47 down vote accepted

You have to sell him the idea. The most powerful sales technique I know is Solution Selling.

  1. Find his pain related to your problem
  2. Come with a solution & facts.

Find his pain related to your problem

Every single human on this planet has multiple pains. Most bosses pain is related to ROI. When they don't get the productivity (and then money) they require, they suffer a lot.

That's your target. You will have to put emphasis on that since this is what you are talking about. Your productivity.

Come with a solution & facts

You must calculate the gain on time a new machine can provide you. There are plenty article online you can find by googling.

Converts the time you gain in money. His pain.

Be cautious about the figures you will send to him. He could use them against you. (see Tim's comment below)

Conclusion

Don't talk about your problem. Talk him about his problem.

share|improve this answer
22  
+1 "Don't talk about your problem. Talk him about his problem." and pain analogy. – Amir Rezaei Dec 16 '10 at 8:30
1  
+1 The "algorithm" works! – duros Dec 16 '10 at 8:36
4  
Just don't oversell yourself in doing so, or it becomes your problem :) – Tim Post♦ Dec 16 '10 at 10:06
Excellent point Tim. I edited my answer. – user2567 Dec 16 '10 at 10:11
1  
Also get him to confirm that he wants that problem to go away. Do it in a leading way (but not obviously leading). "I'm assuming you want me working at peak efficiency?" His answer can only be "yes". If he tries to wriggle out of it, point him back to his earlier answer. – JohnL Dec 16 '10 at 10:19
show 10 more comments
up vote 2 down vote

Do a calculation about how much productivity is lost because of your slow machine. Multiply with your salary and show him that the ROI (return of investment) time for a new machine is 2 months. Add "better motivation" and "better response times in case of emergency" as reasons to buy you a decent machine.

share|improve this answer
up vote 4 down vote

I asked a similar question over on StackOverflow:

It is of course specific to Visual Studio. But the idea (the accepted answer in that case) is that you come up with some numbers to which management and non-technical folk can understand. Measure slow machines as well as fast machines in your company. Hopefully they won't all be slow.

Depending on the size of the company, your boss (unless he/she is CFO or in charge of finance) will need to give a reasoning to his/her boss behind purchasing new equipment.

What you want to show is the time or money it costs the company by not upgrading to faster machines for developers. Idle time, time waiting for the computer to compile a solution, to return data, is generally what you'd want to look for.

If you're idle for 30 minutes, 1 hour, or even 2 hours per day waiting for your workstation to finish something. It adds up over the course of a week, month, year, etc. You will want to plot these out.

Also, avoid using the term "me". It's not about "you" to a company. It's about the company. If things go well, may end up getting a new machine, and your boss will get credit for saving the company x amount of money.

share|improve this answer
1  
this justify the acquisition of an (expansive) SSD drive alone. – user2567 Dec 16 '10 at 11:22
1  
@Pierre 303: I changed job in the summer and my new boss was keen to get me the fastest bit of kit around. I had to talk him down on things to stop wasting money - my development does not really require cutting edge graphics cards. But I did have to insist on a really good SSD drive; makes all the difference. – Orbling Dec 16 '10 at 16:10
@Orbling: Jeff Atwood published some stats on his blog recently. Also ideally, the boss should leave the developer choose his setup. – user2567 Dec 16 '10 at 16:14
@Pierre 303: Whilst I agree, I find most people tend to over do it and end up working on something that is not far down from a supercomputer, for what is essentially text editing and file transmission. – Orbling Dec 16 '10 at 16:31
1  
@Orbling: this is probably not necessary either ;) xtremecomputing.co.uk/images/review/A.C.RyanMorf/image061.jpg – user2567 Dec 16 '10 at 16:33
show 1 more comment
up vote 16 down vote

I just sent my development manager an email the other day where I timed how long it took me to restart my computer, as well as things like opening Visual Studio, compiling, etc. Then I multiplied this by an estimate of how many times per day I do these things, then by the # of working days per year, then by our number of developers, and then finally by our normal costing rate for developer time. The dollar amount of lost time was compelling enough for him to look into getting us better machines. Nothing has come of it yet, but a man can dream...

share|improve this answer
That’s the problem. You never know how to interpret that. – Amir Rezaei Dec 16 '10 at 14:04
2  
Kudos for doing the research and "presenting a business case" for management consideration. A CI server may be the result, as there's more than one solution to some of these problems. – JBRWilkinson Dec 16 '10 at 16:46
1  
We have a CI server. It isn't the cost of doing frequent integration builds. It is the cost of just doing normal development work that I was calculating. – RationalGeek Dec 16 '10 at 17:35
up vote 0 down vote

Buy the new one based on what you want using your money first. Do best your job, show the result to your boss. If your boss gets profit from your job, ask him to reimburse you.

I usually do this in my lab. :-) Very fair, isn't it?

share|improve this answer
up vote 0 down vote

Do a simple trial. If you have some tasks that take a long period of time, that are repeated by other members of your team, show your boss the numbers, and have one person get a real machine. When that person is able to do so much more, then use the data from that test to prove that a new machine for everyone would be beneficial!

share|improve this answer
up vote 1 down vote

If you work at a small company or if the manager sometimes sits beside you to watch what you're doing, or if you get the opportunity to show him/her something, genuinely show how much time something is taking that shouldn't take so long in the first place. Subtly complain about the machine. If you do it right, the person might "feel" it's time for a better computer. My friend did that once and he got new monitors, keyboard, etc.

share|improve this answer
This is pretty much what I did. It only took a couple of years though. – Skizz Dec 16 '10 at 17:08
up vote 2 down vote

A lot of the answers here relate to making a business case for buying you a new workstation. By all means, make that case. But that shouldn't be the primary argument you should be making to your employer.

The primary case should be "I'm frustrated because my computer is so slow". And I'm not just saying that out of some ideal that managers need to cater to all of their employees' desires. I'm saying it because that's the best business case you could possibly make. In the grand scheme of things, the cost of replacing a developer who (presumably) gets things done is by far greater than the $3-4k that they could have spent to get a new workstation. Besides that, what percentage of your income is a new workstation? Hopefully, it's a pretty low one.

In other words, the cost of finding and paying good software developers is the greatest cost any software company has to pay. Buying those programmers the tools they need (or even the tools they want) is tiny in comparison.

share|improve this answer
up vote 2 down vote

I have long since given up trying to convince people about the amount of computing power I need to do their bidding. I will do things with their equipment for a short time, and then what I will often do is bring in my own system (usually, my desktop from home) and sit it next to the computer that they have. Then I sit them down and I show them everything "normal" on their system, and again on my system. My system only cost me about $500 to build, piecemeal, and it runs circles around everything that I have seen in a business environment, even a technicial business environment.

It's a sales technique, of course: the Kirby people (you know, those people that come to your home and demo their "home cleaning system"s to you) call this "killing the old machine," and I have found that at least when it comes to the argument for a more modern, reasonably powerful workstation, or for dual-displays, it is pretty easy to "win" by "killing the old machine," which is really about killing the perception that "everything is fine because it's not broken and/or burning up in a ball of flames".

share|improve this answer
@Michael Trausch If work at a corporation you may have difficulty to connect or bring your own computer. – Amir Rezaei Dec 16 '10 at 18:39
(Essentially I am saying, let the data speak for itself: if the amortized cost savings does not affect their thinking, nothing will, and it's pointless IMHO to remain in pursuit.) – Michael Trausch Dec 16 '10 at 18:40
@Amir: Given the relatively high security of my systems vs. those on every major corporate network I have ever been on, it has never been an issue for me---and the results have always spoke for themselves in a positive manner. Typically it's best to also know the IT management and be "on their good side," as well, because if they can see that you know what you're doing and are willing to vet your machine (security always being the main issue, and something I practice compulsively when it comes to work), you stand a good chance of helping them, too: they will also benefit if you succeed. – Michael Trausch Dec 16 '10 at 18:46
1  
I agree. But big corporation are blind for such efforts. – Amir Rezaei Dec 16 '10 at 18:49
+1 Since in small companies it may work. – Amir Rezaei Dec 16 '10 at 18:50
show 1 more comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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