Slashdot Mirror
Establishing an IT Budget for a Small Business?
tirthas asks: "I am the Information Manager of a small (20 person) architecture firm and have recently been asked by my employer to prepare a technology spending budget. While I have a good handle on what I would like to spend vs. what I must spend, I am having some difficulty establishing a justifiable budgeting method. I have seen examples of 'per employee' methods and 'percentage of revenue' methods, but the dollars and percentages vary widely. What methods do you use to establish your departmental/company-wide budgets, what are your monetary amounts or percentages, and what successes or failures did you have in establishing your budgets?"
Per employee and percentage of revenue are great metrics to meausre your technology spending against other companies, but I wouldn't use either of them as a budgeting tool.
As for getting a budget approved, may I recommend presenting things in a menu-like manner along with an explanation of the benefits of making the investment and thre risks of not making the investment. ~
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
First, create a "consulting" line item in your budget.
Second, put $20 million in the consulting budget for next year.
Third, hire me as your consultant for one year.
Fourth, at the end of that year, I'll tell you how to budget your IT operations.
is what you should shoot for if you listen to idiots like Mr. Mott. HP is in for a world of hurt. They already cut 14500 jobs. If you are making money and employees are happy and you spend 50% of your gross on IT who gives a crap?
Wow... this is a whole can of worms. You're not going to get an appropriate answer on semi-anonymous web forum. I operate my own technology consulting firm who specializes (coincidently enough) in professional service firms such as yours. I earn my living, in part, by answering these types of questions.
I'll need to know the current technological state of your company. How close to capacity is your IT dept running, both in manpower and equipment/services? What are your company's growth expecations over the next two years? 5-10 years? In what role does your company see the IT department, cost center or profit center? In the case of the former, how might we turn that around? How does your utilization of technology come to others in your industry? Etc, etc...
Take the sound bites that you're going to get here with a grain of salt. You're going to need answers that are specific to your business itself. I know it's probably not what you want to hear, but you really should call in some outside help so you can learn how to do it right. Patching together piecemeal advice might cost you your job in the end.
Entrepreneur : (noun), French for "unemployed"
Make a spreadsheet of all expected/likely purchases for the next year. Give rough min and max ranges for quantity and price for each line item. Include things like new systems, servers, upgrades, replacements, software licensing, network hardware, the unexpected, etc. Try to keep IT wages and contract work seperate from hardware/licensing costs. Then use the spreadsheet totals to figure out a good range of expected costs for the year.
Whatever you do, don't forget to budget for ample supplies of chicken blood and other black magic staples. You know you will need it when the boss asks for those apps and hardware that magically do things that aren't possible in the natural realm.
At my company (14 employees-ish) the budgest for IT is just "Find out what we need, buy it, and give us the reciepts". Seriously. Our IT guy just does what he has to, and justifies it when asked about it. If he can't justify it, he either gets limits on what he can spend next month, or it gets returned for something cheaper. With small companies with extremely variable budgets (months our company actually get PAID result in much higher IT purchases), it's best to take it one month at a time, and document reasons for EVERYTHING!
I like to place meaningful quotes in my sig, so people will know that I know what meaningful quotes are.
That's an important question, because if they've actually hired you to be ``Information Manager'', they're going to have entirely different expectations than if they've hired you to be a draftsman and dumped this extra responsibility on you.
If you're a draftsman with an extra burden, I recommend that you look for the thing(s) which will let you solve a few small problems, give you no new problems, and not waste any of your time on adminstration. Find the price tag, and you have your budget. If the number is too small, new machines all around (or, just for the partners and their favorites, and let their old machines trickle down). You can't afford to neglect the one part of your job they understand (the drafting), so don't let yourself get trapped in system administration!
If you were hired to be a full time IT manager, why are you asking us for advice? Figure out what they need, tell them what it costs and how it will save costs and increase revenues. You do know how to do that, right?
See what I've been reading.
I too work for a small company that does not pull a lot of profits. The cash flow is very seasonal and the budget is kind of mystical.
What we do is just sit down and assess who needs new computers that year. How much you spent last year on incidentals (CD's, cables, RAM, etc).
A rough guess is to add 20% to what you have down on paper. Of course you can only make an accurate budget knowing your past spending habits and what things are on the horizon.
One thing of note. I worked for a big hospital a few years ago, they had a multimillion dollar budget that grew every year. The mentality was that you had to use it or lose it, so the budget always grew. Some guys (including myself) offered some F/OSS solutions that would have save many thousands of dollars. We were soundly rejected. When I worked at Discover Card, they gave incentives to those that could save money, not tell them to take a hike.
Anyway, for a small business like mine, just make a list, look back at your previous purchase orders and see what history tells you.
This space available for rent.
I hate to say it, but a lot of your spending will depend on your clients. If your clients run Autodesk CAD software, then so should you, and that will make a huge dent in your bottom line. Our clients stagnated on AutoCAD 2000 for years, then just this month decided (and these are fortune-500 retailers, mind you) "oh, lets upgrade to AutoCAD 2006, so should YOU"...
I'm the sole IT person in a 50-person architecture/construction management firm. Our spending varies year-to-year but there are a certain amount of annual expenditures. Things like antivirus software (both at the server and desktop level, CAD upgrades (we stagger our and are getting onto Autodesk subscription... it hurts in the beginning but pays off after a couple of years).
I do my best to save on the software side and apply the savings to better hardware. By this I don't mean go out and pirate what you can't afford, I mean look for OSS alternatives to things you may think are a must-buy. I run sendmail with Mailscanner/spamassassin on FreeBSD and linux, instead of Exchange. I run supplemental Samba servers on quality HP servers. I do run a windows domain, because its just easier for me to manage than a Samba-based domain. Believe or not, MS stuff just works in my organization. Our industry-specific accounting software is windows-only also, so we're kind of stuck there.
Like someone before said, you need to take into account your corporate growth goals and decide if you want to invest in a lot of good hardware that should last a long time, or buy cheap initially and replace things when needed.
You need to understand what your business does firts... what applications does it use, what sort of workflows, types of documents produced, versioning requirements, etc. and then look at the system you have and see where the holes are.
I have done this for myself (started back in 01/2003) and also for clients (as a side) and to be honest, for a small business like you talk about, there is not really an IT budget. The point is to spend only exactly what you need to do the job, and if additional benefits can be realised for no extra cash, then that is a bonus.
Any fool can talk, but it takes a wise man to listen.
If you are going to be purchasing any proprietary lock-in software. To budget in for the enevitable operating system upgrades and office upgrades. Then of course there are the AV software subscriptions and so forth.
Do careful planning and always have a back-up plan. What happens if the vendor you bought software product X from goes bust - what happens to your support?
Electronic Music Made Using Linux http://soundcloud.com/polyp
Don't stop with PC's and applications. Include phones, Internet access, printers, faxes, document repositories, paper, toner, CD's, you name it. If the business strategy is best supported by quill pens and parchment paper, though, don't be afraid to go low-tech.
The partners probably have an idea of how much they want to spend on you, your team, and what you and your team do. Find out.
Do not trifle with that step. If they think they can afford $30,000 a year on technology and you present a $300,000 budget, the next sound you hear will be the axe falling.
Never forget that your firm is in the business of architecting things. Having cool computers with the latest applications that do not contribute directly and measurably to the architecting of things is deadly.
First, ignore all the per-capita methods. They won't work.
Now, take your lists of what you need to spend and what you want to spend. Lay out the items, give each one a priority ranging from "must have" to "would be nice". Provide a justification for each one and for the priority you gave it. Give all justifications, not just the best one, with examples from reality (eg. the justification for the anti-virus software might make reference to actual virus infestations in the company in the previous year and how much they cost in money, time and resources). Be sure to account for recurring costs in future years (eg. the service contracts on equipment, anti-virus update services, etc.).
Once you've got your list, sit down with the Finance guys and figure out how far down the list the business can afford to go.
Realistic budgets aren't based on per-capita expenditures or percentages of revenues or profits, they're based on what the business actually needs to spend to stay in business.
First, ask the controller or CFO what the company uses for a standard method, such as internal rate of return or payback period. You don't want to walk into a presentation with numbers using method X when the management thinks in terms of method Y.
Here's a 9 point, back of an envelope plan which sums up what the big consulting firms will tell you (in less words).
1. Review the Business Plan. What does it tell you? You expanding? You going all e-business? What? Wishes as well as reality.
2. What about trends in the architecture world? Trends in technology? Cheap storage, cheap bandwidth etc?
--> Try to work out what capabilities are suggested by all of this. eg - if your firm says they want to go e-billing then, hey, thats a capability. Show these to your firm and get them to rank them, score them whatever.
3. You haven't just started up, I suppose, so how does your current IT match up? Do a big list of capabilities you have identified and assess each one against your current apps/functionality.
4. What would a future IT system look like if it were to deliver all these capabilities? How could you leverage some of the technology trends you identified in to creating growth or profit? (invoices by pdf = save on printing and mailing etc. Recruit new staff just via the web = savings etc )
5. Can you decribe an optimal IT applications and technology architecture yet? Would that be an open accessable database feeding billing, web portal, collaboration, etc? Web based? Accesible in the move, laptops etc?
6. Now do a gap analysis of what you have versus what you want. Describe how you could move between those two states. Ie projects, dependencies etc - migrate this database, use a temporary TS server, then move to app x etc etc. remember culture (saving local vs contributing centrally - loss of power for the individual) - have a plan for selling culture changes too.
7. How could you pay for this - what levels of investment would it take? How will the changes be paced - what happens this year and what next. Will there be an investment phase of a few years then a leveling out. Talk business cases here - ROI.
8. Map the risks and critical success factors. Map your communications plan for the changes.
9. Present. You now know exactly what you want, why you want it and in terms that the rest of your staff team will understand - they nedd the capabilities not the software itself. Easy sell really.
We go though this every few years and it is an effective enough methodology.
(Support staff salaries) + (# of PCs hardware cost) + ($699.00 per user to SCO) = your budget.
Trolling is a art,
Friend, I'll give you credit for actually asking the question. It's a smart person who admits when they don't have the info they need, and tries to find out the answer instead of winging it. But, you're going about the solution all wrong. Proper IT support is not just about numbers - there's soooo much more.
First, you need to get a handle on what you're employer actually expects out of it's information systems. Some firms want a glorified typewriter, some want it to do most everything. Some can't stand spending a dime on a machine, others only want the latest and greatest. Guaging the culture of an organisation is paramount.
Then there's risk. This is, IMHO, the biggest factor in actual budgeting. How much risk to the companies bottom line will any expendatures entail? How much risk is involved with doing nothing? How much risk is involved in your compeditors advancing further in IT efficiencies? You need to show that what you intend to spend will protect current operations, or there won't be a budget for anything, let alone IT. As well, if you can show that there's significant risk in not spending a certain amount, most CxO types will approve what you're asking for.
Then you can get into the numbers, and justify your expendetures with ROI.
OK, the problem? My experience lately has been that professionalism is sorely lacking in a lot of IT departments. I've heard people lament "They're driving the company into the ground anyway - why should I care?", or justify an over priced solution on what other departments get, while bitching about how under appreciated the IT department is. Your department is under appreciated because it isn't communicating properly with the people you serve. It's that simple.
We techies tend to look down on business users (they are just lusers, after all), but they're the people who drive the ship. How they're driving it should not matter one whit to you effectively doing your job. Hell, maybe some information that your systems provide will turn into a cluebat and wake up a sleeping bridge. You need to think a bit like them in order to effecitvely communicate what you're trying to do and why. You need to use thier terms and jargon, even the (YECH) buzzwords. This seems to be anethemia to a significant portion of the Slashdot crowd.
A professional, when he accepts a job, will do what's right for his customer within his area of expertise, no matter his own personal opinion on the customer or the customers direction.
Soko
"Depression is merely anger without enthusiasm." - Anonymous
There is a point to this (honest): IT often has to play in cost-benefit analysis against other corporate budgets, including facilities.
Liberty you never use is liberty you lose.
Your firm must be big enough to have an accountant, if so, time table a long meeting with him/her/it and thrash it out with them. You might as well, it'll only end up on their desk again anyway. Come up with a range of options that meet the requirements and the account will help you put figures on things, including your assessment of the liabilities in el cheapo solutions. Personally, I've had a good success level and good experiences with turning the "gatekeepers" into collaborators that way.
Plays violent online games as: Nerfherder76
I imagine if you're in charge of administering most of the daily tasks (backup, network maintenance, upgrades), there is no labor cost. This makes it very easy for you.
I would put something together where you talk about the existing infrastructure, and maybe some proposed changes. Rank these changes in priority from "Low" (we can live without them but they MIGHT provide efficiencies) to "Medium" (realistically these will be needed, provide a good benefit for the investment) to "Musts" (these have to be done in the next 12 months otherwise the business will not function as it does today).
For example, "Low" might be buying larger, better displays for the office and a plasma TV for the client presentation room. "Medium" might be to buy larger, better displays for some of the drafting people, if the current ones are already a bit old or unreliable. "Musts" might be the monthly costs of internet/voice service (make sure they're not double- or triple-counting anything) or replacing a monitor that you KNOW will be gone within three months.
Also, make an approximate tally of the current value of items in the office, as far as IT is concerned. What if a file server went down? How much is the replacement? How much would a new bells-and-whistles replacement be?
I would probably draft up a "recommended" budget:
Recommended Budget: Medium
Then, above and beyond that, ESTIMATE the "Oh, crap" budget, for example:
Clearly spell out that the first part is "in your control" and the other part is BEYOND your control. You may find that the "in control" budget is $11,000, while the "beyond control" budget is $25,000. Or the "in control" budget is $35,000 and the "beyond control" budget is only $7,000. Approach each of these scenarios differently.
Once this is all done, write a half-page summary on the front page. One paragraph explaining the "in control" budget and a realistic range, plus the benefits. Then a second paragraph with the "Oh, crap" stuff.
I'm making the assumption that your employer wants to know the following:
Succesfully and accurately answering the first three questions will make the fourth one easy.
Small potatoes make the steak look bigger.
Hi-
I'm the sysadmin at a ~25 person architecture firm, and an architect, too.
What I do for budgetting is to start with an inventory of every piece of equipment we own. Then I assign an approximate lifespan to each thing, based on experience. Couple that with when each item was purchased, and replacement times and potential budgetting scenarios start to emerge. There are a couple of policies I try to wedge in there to steer purchases, such as I like to put the new machines in front of the heaviest, most demanding users, which for us is the more billable staff. I'm forunate in that I have management's backing, so that there's never any whining from project managers when an intern gets a kick ass machine, and the PM gets the hand-me-down. By keeping the machines coming in, I can keep everyone pretty happy. In addition, whenever possible, I try to upgrade our software versions either every two or three years, on everything except the Microsoft hegemony, which I only upgrade through churn of new workstations.
Around the fairly regular annual purchases, I then stategize the big ticket items: new plotter, Autocad updates, expanding licenses of Photoshop/InDesign, implementation of a new accoutning system, or what-have-you. This helps even out the costs from year to year. None of it's rocket science; you just have to put your head into it and figure out what it would cost to maintain your company's current level of technical prowess (x machines per year, etc), and that what it would cost to further develop the skillsets (by getting new software, and doing more training).
FWIW, we spend ~$45k per year, which works out to be ~2.5 - 3.5 % of revenue. My bosses have never gone for a straight percentage method, but it's my preference, since it automatically adjusts itself for good times and bad. Instead, I develop a budget, they either say yes to the whole thing, or specifically exclude one or more items. Then I establish the priorities and give them an idea of at what points in the year I'd like to spend money. They coordinate that with the company's cash flow, and if the year turns sour, I just get reigned in on my spending.
Sorry if I'm rambling too much. OTOH, if you want more, post a response, and I'll give you my email address.
Joel
responses by IT insiders who really have no understanding of the business picture.
You want responses from people who understand the small professional services firm.
My experience as an IT fence sitter in a small professional service firm is that if you were to recommend it, as the expert in the firm, the principals will be willing to pay for it.
I would simply make a list of your expected needs and spread it out over the best time periods to implement the changes.
If your owners are "cheap," don't try and be "strategic" into manipulating them into any more than they need. If they want the least expensive solution then your job is to find and make the least expesive solution work.
If they think throwing gobs of money at the systems will guarantee zero problems, then make sure and keep their expectations in check.
I like to point out that 4 hours of downtime costs X amount of employee time in $.
We use an outside consultant who we have basically full confidence in, and we simply just follow his suggestions.
About a year ago, my company's IT department rolled out LCD screens to every employee and I was ecstatic. Not so much because I had been clamoring for a new screen (I hadn't even thought about it), but because it was an upgrade that actually made some kind of logical sense.
Think about it. What I do all day is mostly send e-mails and work in Microsoft Office. A traditional 'upgrade' would be to drop a new, faster computer with a big hard drive on my desk. But not only does that take a lot of work on behalf of IT, not only is it incredibly disruptive to my workday, but it doesn't really benefit me at all. I can probably store my entire work folder and all its accumulated contents since I began working here on a single USB keychain drive. My CPU needs were met and exceeded some generations of hardware ago.
On the other hand, a new LCD screen that's crisp, clear, and easy to read -- as opposed to some legacy, piece of junk CRT that's been getting blurrier and dimmer for years -- is something tangible that I look at and notice day in and day out. You might think it doesn't result in an increase in productivity, but I disagree completely. In fact, if someone had given me the choice between the monitor and a new CPU, I would have taken the monitor in a second.
Breakfast served all day!
The other thing to keep in mind is that you need "thoughtful" input on what you are buying. Its one things to decide if laptops or desktops would be more cost effective IT wise, its another to ask people if they prefer a laptop or desktop but the real question is, what does it mean for your culture and work style.
Based on what tools people are given they will find different ways to work. In the most basic example, give them laptops and they will work in groups but also from home more. Give them desktops and they will come in but perhaps not be flexible enough. This applies to quite a few hardware and software purchese. When you save $1000 dollars by giving the two sales guys slower machines are you saying they are not as important as the 20 developers? Perhaps you should just spend the extra $1000...or perhaps not.
I only suggest you keep in mind that even decisions that seem totally IT or budget based can have wider implications.
Someone at a management level should be thinking though these questions as part of this process.
is often one of the easiest. The bean-counters, especially the computer-illiterate ones, look at the WidgetMaster 2000 and all they see is a price tag, because they have no idea what this magic device does, or how it is going to help the company make money. When all you have is a price tag to work with, all you can do is go cheap. ;)
So, for any item you would like to have, (I didn't say "have to have") you need to put together a simple description of the item. Explain in plain terms what it does, this will help the people that have half a clue. Then spell out how this item will save money or make money, in simple terms. (go on the assumption that what you have just told them is ALL they know about the item at this point) Explain how long the item will last, and how long it will take to break even. Explain how much money you roughly estimate the item will make your company over its lifespan, factoring in its initial cost and the cost of updates, maintenance, etc. Don't cut corners on the future costs - they need to know if you have to buy a new license for the software every year or renew a service contract.
Do this for anything you would like to have. Do not restrict yourself to things you think you will get, or just the things you find essential. Include it all. They are unlikely to approve everything, so there is no sense in being picky about what you ask for - you can't get what you didn't ask for. You might drop a hint to the reviewers when you hand it in, giving them a hint as to how much of this you expect to be approved. This will ease tensions as they may still not understand how much of this you need and how much you want. If you tell them you're hoping to get at least 50% of what's on the list, they won't freak out trying to figure out where they're going to get budget to get 95% of your list. (this usually leads to you getting a completely random assortment)
Depending on who's evaluating the list, it may also be useful to break the items down into groups. "Essentials", things we need to get to stop hemoraging money. "Needs", things we need to become more proffitable, improve efficiency, improve customer response time, etc. "Try-outs", things we'd like to get into and we're looking for a sample of the technology to see if it's worth a buy-in. "Extras", things we strictly don't need, but that may have a positive effect on the company... these can include things that simply make employees' jobs easier to do or more tolerable. (how about a radio for the mailroom?) If they're short on cash they may very well go entirely on this grouping to determine what they get - maybe they only get the essentials and the needs this year.
This does a lot for you. It stops them from buying stupid things you don't really need rather than the things you had to have yesterday. It also helps them to make informed decisions about what they should buy and what can wait until next year or next quarter. And it helps you because you can push the tech in the direction you are prepared to go rather than getting a spray of differnt items which could take your people in three different directions at once. You are their only source of information right now and what you tell them is very important to your business. You aren't actually the one spending the money, but they are depending on your professional opinion right now to spend that money where it counts. What you tell them will determine the direction your I.T. goes for the next 10 months.
I work for the Department of Redundancy Department.
As we all know, the biggest dent in budgets is made by personnel costs.
Unless the information manager is only doing information management 2 or 3 hours a week, I would say it is useless for such a small company to have one on staff. If the company has such a strange spending pattern that spending on other things actually outweighs hiring an information manager, then you're not doing your job; and if you are doing your job, your job is redundant.
This is from the perspective of an IT geek in a (i'd call it a small company as little as 4 years ago) medium size company. Of course I'd like to take all the credit for the company's growth--many others contributed significantly in their own areas of responsibility.
The problem with consulting is that managers tend to look at IT spending as buying a commodity. Spend more, and you will get more of this stuff called IT. Consultants make it easy for managers to not have to understand how IT impacts business processes. Sure, it is the consultant's job to recommend those systems, but without an IT aware person on staff, how does the manager practically evaluate the consultant's recomendations.
As a director of IT in my company, I find myself spending a large amount of time educating other managers about what IT can and cannot do. Again, consultants can do this work, but they have a profit motive to spend more and do more on IT projects. The fact is, consultants love to keep their clients in the dark so they can be milked for more money. A consultant can never understand the inner workings of a company, because unlike the IT staffer, they don't see the day to day.
Consultant bashing asside, it would be tough to justify one full time IT person for an office of 20 people in some, but not all offices. Though, achitecture firms are the obvious choice for full time IT person since that IT person can do other design oriented things as well--web, graphics, etc .
Our company has looked at other comanies in our industry and have found that bigger companies in particular spend a lot more on IT when they rely on consulting. I guess they never internalized how to use IT before they got big. End result--we are a lot more efficient than big companies.
Outsourcing does have a role. Don't be afraid to outsource those areas your IT person doesn't have strength in. Only pay to outsource those areas you need to--that way you don't have to pay consultant rates for someone else to do all your IT thinking for you. As your company grows, you can wean yourself away from consultants as you complement skills of existing IT staff with new hires. Honestly, besides some design oriented stuff and some basic office network stuff, this architecture firm doesn't need a lot of high end IT talent. When you have multiple offices, hire a good networking/telecom guy to tie all your offices together.
Here's a tip that often gets lost on people: Everything in business spending -- and by that I mean EVERYTHING IN BUSINESS SPENDING -- comes down to one thing: Return On Investment. How much do you get back for what you spend, and how fast?
Of course, quantifing things like worker productivity, morale, and community image can be hard, but you can make a good go at it. Look at how much time people spend waiting for the computer, bitching about the computer, or otherwise Not Getting Stuff Done Because Of The Computer.
It's amazing how many people don't get this. User: "I want this, and this, and this." Me: "Will it help you do your job?" User: "No." Me: "Buh-bye now." Salesdroids, too: "Our new Plasmomatic 6000 SUX can make copies and solve the halting problem!" Me: "But I just want to make copies. Why should I pay for more?"
More serious examples: When evaluating new network printers, I look up how much paper we bought for the thing over the past couple years and figured out average pages per month. Cost-per-page is easy to find in spec sheets these days. I picked a few models, got the costs and calculated differences in same, and then figured out how long the savings would take to pay for the price differences. This pointed right at the model to buy.
I found the QA people were doing diagrams for their procedures in, $DEITY help me, Microsoft Paint. Getting approval for a vector graphics program was simple once I showed how much faster it would make things.
Conversely, when the new CAD guy wants a high-performance laptop but will only be on the road three times a year, it's easy to point out that the ROI is not there. He gets a much cheaper high-performance desktop, and he can limp along with one of the "floater" laptops if he has to.
If an investment can't pay for itself, you don't buy it.
A lot of budget decisions become a lot easier once you understand this concept.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
i think the real question being posed to our friend is: for our small architectural firm, how many copies of autocad photoshop illustrator are we buying this year? what is the minimum software and hardware we need to purchase to keep our employees productive whitout breaking the bank? i was speaking with a guy from a small architectural firm last year during a focus group, and what i gleened from that conversation is how much architecture relies on presentation. marker renderings and topographical models have been replaced by 3d renderings and photoshop compositions from existing elevation photos...
at the end of the day it's a head/seat count for hardware and software. i usually count on at least 1 major upgrade a year which i'll be asked to purchase at a cost of 30-70% of original cost, plus 1-2 major bug fixes, and 1-6 minor fixes at little to no charge, regardless of software. i also expect production hardware to be replaced(desktop computers, large format plotters, centralized workgroup laser printers, color copiers) every 3 years, and system hardware(servers, raid storage, network equipment, tape jukeboxes, server archive software, other server software) to be replaced every 5 years.
of course YMMV in your particular scene, so don't ask your boss to replace a bunch of gear if he's a cheapskate. pose every need as exactly that, a need. we can't do our presentations WITHOUT photoshop CS2, we NEED to upgrade 10 copies of autocad to version 2006 our senior architect is complaining that his BOOTLEG copy of version 14 is taking a long time to launch, our HP5000 is breaking down and the cost or repair is MORE that a years worth of lease payments, if we don't buy X copies of Y software, one of our disgruntal former employees will drop a dime on us and call the SPA or BSA, so here's a schedule for how to get legit in the next 3 years... you get the picture... if you can't buy it this year, schedule it for next year, unless it's a break fix.
three can keep a secret, if two are dead - benjamin franklin