MORE ENTRIES
- Whether to mention a pregnancy in a job interview
- A possible meeting protocol
- What are an end-user's responsibilities?
- Another take on opening PCs, or not
- Getting some process going
- Selling a more open environment to management
- Running an effective meeting
- Licensing rules for virtual machines
- The ROI of metrics
- Legal challenges to virtual machines
Advice Line | Bob Lewis »
Penetrating the mysteries of IT budgeting
Dear Bob ...
OK, I like what you have said and the questions you have raised concerning employee compensation (in "Poor Joe" and "Comp logic," Keep the Joint Running, 10/22/2007 and 10/29/2007). I have a related issue you may like to consider.
It is that time of year, again, in most organizations, where we need to develop next year’s budget. Like many other companies this year, we need to look at ways to cut expenses - at the same time that we (in IS) are being asked to tackle several major, complex projects - involving system-wide impacts, multiple interfaces, all users, and not to mention continuing support of our existing applications.
Sound familiar? The list of projects would lean towards the addition of staff, not reduction or maintenance of status quo.
The conventional logic is that "census" is down, revenues are down, and therefore we must reduce expenses: translation - no more staff.
In IT, the workload is not a function of the in-patient or out-patient census. Our workload is a reflection of such things as:
It may be a useless endeavor, but it is very difficult to talk numbers to Accountants without some number.
So – thoughts? Ideas? Any research out there? Have I snorted too much coffee, resulting in total malfunction of my two remaining synapses?
- Budgeting
Dear Budgeting ...
Just my opinion: You'd be better off drinking your coffee than snorting it. Tough on the nasal septum. Trust me on this.
Just my opinion on the question you asked: You are, for the most part, on the right track. Your next step is to segment IS, because different areas of responsibility have different staffing drivers. So:
Major projects: Staffing is purely a function of your company's IS Governance process. The more projects they approve, the bigger they are, and the more they want to have happen concurrently, the more staff you need for them.
Discretionary enhancements: Similar answer, only governance for discretionary enhancements shouldn't be the same process. I usually prefer to handle these as a budgeted allocation, but instead of having a dollar budget, department heads receive an allocation of developer hours. The IS Governance committee (or whatever) is responsible for establishing this allocation. The more enhancements they want to have happen concurrently, the larger the budget.
Non-discretionary maintenance: It's non-discretionary. Inform the IS Governance committee of the staffing required for this. To take the sting out of future staffing requirements, make sure every project's business case includes the impact on maintenance staffing.
Discretionary infrastructure: This category supports infrastructure build-outs for new facilities, remodeling efforts, departmental moves and so on. Let business plans drive it. If the answer is, "We don't know what will be required next year," it's perfectly fair to respond, "Then it's awfully hard to predict what staffing we'll need to support it. For now, I suggest we leave last year's staffing alone, assuming we'll have to deal with the same level of requirement this year."
IS Operations: This is where servers per sysadmin ratios and that sort of thing come into play. I'd advise including a continuous improvement target in your ratios - as staff get better, processes improve and you improve your infrastructure, each sysadmin should be able to handle more servers.
End-user support: This is the biggest challenge. Start with the current ratio of end-users to analysts. Then propose three alternatives - one based on reduced support, one supporting at current levels, and one that invests in maximizing business user effectiveness. There are severe limits on the possibility of continuous improvement in this area because most of the improvements that would let you increase the end-user/analyst ratio depend on improvements in your end-users, not your analysts.
I've probably left out a few areas of responsibility, in part because they depend on how you're organized (do you have a separate Architecture function to house your DBAs/Data Analysts, for example). You can apply the same sort of thinking to these as well.
- Bob
RATE THIS ARTICLE:
November 07, 2007 | Comments: (0)
Penetrating the mysteries of IT budgeting
Dear Bob ...
OK, I like what you have said and the questions you have raised concerning employee compensation (in "Poor Joe" and "Comp logic," Keep the Joint Running, 10/22/2007 and 10/29/2007). I have a related issue you may like to consider.
It is that time of year, again, in most organizations, where we need to develop next year’s budget. Like many other companies this year, we need to look at ways to cut expenses - at the same time that we (in IS) are being asked to tackle several major, complex projects - involving system-wide impacts, multiple interfaces, all users, and not to mention continuing support of our existing applications.
Sound familiar? The list of projects would lean towards the addition of staff, not reduction or maintenance of status quo.
The conventional logic is that "census" is down, revenues are down, and therefore we must reduce expenses: translation - no more staff.
In IT, the workload is not a function of the in-patient or out-patient census. Our workload is a reflection of such things as:
- Number of Users
- Complexity of the System (number of interfaces, etc)
- Number of devices to support
- Number of applications to support
- You get the idea ...
It may be a useless endeavor, but it is very difficult to talk numbers to Accountants without some number.
So – thoughts? Ideas? Any research out there? Have I snorted too much coffee, resulting in total malfunction of my two remaining synapses?
- Budgeting
Dear Budgeting ...
Just my opinion: You'd be better off drinking your coffee than snorting it. Tough on the nasal septum. Trust me on this.
Just my opinion on the question you asked: You are, for the most part, on the right track. Your next step is to segment IS, because different areas of responsibility have different staffing drivers. So:
Major projects: Staffing is purely a function of your company's IS Governance process. The more projects they approve, the bigger they are, and the more they want to have happen concurrently, the more staff you need for them.
Discretionary enhancements: Similar answer, only governance for discretionary enhancements shouldn't be the same process. I usually prefer to handle these as a budgeted allocation, but instead of having a dollar budget, department heads receive an allocation of developer hours. The IS Governance committee (or whatever) is responsible for establishing this allocation. The more enhancements they want to have happen concurrently, the larger the budget.
Non-discretionary maintenance: It's non-discretionary. Inform the IS Governance committee of the staffing required for this. To take the sting out of future staffing requirements, make sure every project's business case includes the impact on maintenance staffing.
Discretionary infrastructure: This category supports infrastructure build-outs for new facilities, remodeling efforts, departmental moves and so on. Let business plans drive it. If the answer is, "We don't know what will be required next year," it's perfectly fair to respond, "Then it's awfully hard to predict what staffing we'll need to support it. For now, I suggest we leave last year's staffing alone, assuming we'll have to deal with the same level of requirement this year."
IS Operations: This is where servers per sysadmin ratios and that sort of thing come into play. I'd advise including a continuous improvement target in your ratios - as staff get better, processes improve and you improve your infrastructure, each sysadmin should be able to handle more servers.
End-user support: This is the biggest challenge. Start with the current ratio of end-users to analysts. Then propose three alternatives - one based on reduced support, one supporting at current levels, and one that invests in maximizing business user effectiveness. There are severe limits on the possibility of continuous improvement in this area because most of the improvements that would let you increase the end-user/analyst ratio depend on improvements in your end-users, not your analysts.
I've probably left out a few areas of responsibility, in part because they depend on how you're organized (do you have a separate Architecture function to house your DBAs/Data Analysts, for example). You can apply the same sort of thinking to these as well.
- Bob
Powered by ScribeFire.
Posted by Bob Lewis on November 7, 2007 05:38 AM
RATE THIS ARTICLE:
-
- COMMENTS
|
Three books. Three ways to change the world, your life, or at least Bob Lewis' bank account. Leading IT: The Toughest Job in the World distills the world of IT leadership into eight learnable skills and gives you concrete, practical techniques for each one of them. Bare Bones Project Management: What you can't not do makes project management manageable, even for first-time project managers with no formal training in the discipline. ManagementSpeak: What managers say/What they mean … well, it won't help your career, and won't make you a better manager. Mostly, it will make you chuckle, guffaw, and maybe even chortle. Make friends - it's the perfect gift for anyone who has ever suffered through one of those meetings. Order your copies today! |
TOP STORIES
Sun to clarify JavaFX planMS's dev tool service packs
HP in talks to buy EDS
Developers' role shifting
MS: XP SP3 reboots OEMs' fault
Apple: iPhone out of stock
Can Sun rejuvenate Java?
Powerset unveils Google-killer
FBI worried about Cisco gear
AMD updates quad-core Opterons
ADDITIONAL RESOURCES
- Virtualization: A Step by Step Approach to Success
- Dialing up Agility with Business Transformation
- 5 Things You Need to Know About Storage Virtualization
- Is your smaller organization ready for High Availability?
- Is system maintenance doing more harm than good?
- Virtual Test Lab Automation: Manage development infrastructure
