IT Today Catalog Auerbach Publications ITKnowledgebase IT Today Archives Book Proposal Guidelines IT Today Catalog Auerbach Publications ITKnowledgebase IT Today Archives Book Proposal Guidelines
Auerbach Publications

IT Performance Improvement



Networking and Telecommunications

Software Engineering

Project Management


Share This Article

Mixx it digg

Service-Oriented Architecture
Implementing the IT Balanced Scorecard
The Business Value of IT
Managing Global Development Risk
Service Oriented Enterprises

How Much IT Is Enough?

by Michael D. S. Harris, David Herron and Stasia Iwanicki

IT is a strategic differentiator. Often, it is a single force that determines the speed and agility of an organization. Think of it as the vehicle for the journey toward world-class results. It can provide different combinations of safety (quality), speed (high performance), and capacity at different fuel efficiencies. Do you need a Lamborghini?

Do you want one? How about the cost? The question is, what specific attributes do you want or need from this vehicle? Why buy luxury add-ons when base model options will do? What passenger capacity or towing capacity is required?

How long does this vehicle need to last? One year? Three years? Seven-plus years? Decisions about IT spending are a series of trade-offs. The key to making the right decisions lies in first knowing the compelling needs to achieve the business strategy. Establishing the strategic enablers (most critical elements necessary to deliver on the strategy) generates the focus for planning activities to achieve this desired future state: efficient IT, nimble IT, high-quality IT, world-class IT, etc. Performance is defined differently depending on the strategic enablers critical to the business. In Chapter 11, Where Are We in Relation to Industry Peers? we provide performance data for the dimensions of:

  • Productivity
  • Cost
  • Duration
  • Quality

These apply for both new development and enhancement activity. Comparison of these dimensions supports decisions about where to prioritize spending against the known strategic enablers.

To determine how much IT is enough, the scope of the IT budget must be defined. Typically, the IT budget is comprised of some mix of the people, processes, and technology (hardware and software) spending for an organization. The IT budget can be managed as a cost or profit center, functioning through a department, a business unit, an entire division, a company, or a set of such entities that we call IT Providers.

We have defined a number of ways to measure IT value in Chapter 2, How Do I Measure the Value of IT?. Here we present four value metrics to explore how to examine when the level of IT spending may be enough. Each dimension on its own can stand alone; however, together they provide a multi-dimensional picture of technology spending in relation to macro indicators or high-level characteristics:

  1. Return on investment (ROI)
  2. IT spending as a percentage of gross company revenue
  3. IT distribution analysis
  4. Organizational evaluation

Finally, we consider the trade-off between cost containment and innovation.

ROI or Return on Investment for IT Spending

The ROI is as easy as thinking of the return rate for your savings account. You deposit $10,000 at an annual percentage rate of 5 percent. One year later your account is valued at $10,500. Your ROI is $500, 5 percent of the $10,000 investment.

Technology is an investment. The spending provides a return. This is also true for required changes like Sarbanes-Oxley compliance or other regulatory requirements. It represents the cost of remaining a going concern (cost of staying in business) or keeping your CEO on (or off!) the front page of the Wall Street Journal.

Consider the following example: a business unit requests a business application that will create a competitive advantage in the marketplace. That "advantage" needs to be understood and communicated by the business to the IT Providers:

  • How is this application expected to improve revenues?
  • How is it expected to reduce expenses?
  • Are there secondary benefits to be realized by streamlining a process or improving the customer experience?
  • How should this initiative be prioritized versus all other opportunities in the queue?
  • What competitive advantage does this application provide? For how long?

The business states that the CEO designated this as a top priority. It is expected to bring $10MM in increased sales over the next five years.

Now the IT Providers have to estimate the expense to develop the new application. Technology costs will be $2.5MM and take six months to be in full production. (Note that we are assuming an efficient and effective estimation process based on historic data - unfortunately this is not as common as we might expect and hope.)

ROI = (total benefits - total expenses)

ROI = ($10MM - $2.5MM)

ROI = $7.5MM over five years

The results can also be expressed as a ratio: in this case 1:4 representing $4 of return for every dollar invested over the same five-year period. Often, the ROI is expressed in annual terms for easier comparison to commonly used interest rates. Over the five years the ROI is 300 percent. Making some simple assumptions, the annual ROI would be roughly 25 percent. An excellent investment!

There are many other factors that should be included in the expense base such as the total cost of ownership to support and maintain the new application. Factors like these vary for each organization. The most important consideration is to use one method consistently to allow for a true comparison from one internal investment to another.

How much ROI provides enough return or a return soon enough? As we noted in a previous chapter, "beauty is in the eye of the beholder." What return do you expect on a car purchase? The only vehicles that appreciate are the Lamborghini and various antique cars - hardly a good model for an IT system! ROI is a single dimension in a decision of whether to move forward; it affords a direct financial look at the potential for return.

To understand some of the limitations of using only the ROI dimension, consider "Growth Bank" and "Old Bank." Growth Bank sought tremendous organic growth at almost any price. A persistent belief was that spending $1 to make $2 was a good investment. Growth was explosive and organic. It took years for the analysis of total cost of ownership to catch up but eventually it did. IT was very expensive in comparison to peer groups, their applications were built independently, and there was tremendous redundancy in the services they provided.

Old Bank moved at a slow pace, carefully considering every investment. It targeted projects with a two-year payback. ROI had to be at least 5:1. Innovation was rare. Expenses were always toward the lower end of peers. Time to market for new products was always behind industry peers. Growth was through acquisition.

Both examples are approaches of very different, real-industry peers. Both banks are in the top ten, even today. ROI is not a single-threaded answer.

IT Spending as a Percentage of Gross Company Revenue

Defined as the total IT expense divided by the total gross revenue of the organization or company, this includes the entire IT budget, the staff, hardware, software, licenses, contracts, application development, and support. For example:

Company X gross revenues from sales$200 million
Company X IT expenses$18 million
Percentage IT spending =(IT expense / gross revenue)
Company X =9 percent of gross revenue is spent on IT

In our experience, we have seen these numbers as low as 3.4 percent and as high as almost 12 percent. The spread tends to be smaller in specific industries. In an industry like financial services (in a growth year) we would see targets of 6 to 9 percent1 depending on factors such as:

  • The need for new development in technology
  • The organization's reliance on technology
  • Opportunities for consolidation

IT spending as a percentage of gross company revenue is a measure that has been collected and reported for years, so there is a substantive amount of data available by industry and other factors (U.S., non-U.S., CMMIŽ level, company size). Unfortunately, the missing parameter in using this data is whether the industry is in a growth year or a year of restructuring (e.g., lower marketing spending resulting in lower revenues means higher IT spending as a percent of revenue without changing the IT demand). It is still a telling number to calibrate for year-over-year spending. For example, this measure can help paint the picture of a truly neglected IT department which has flat IT spending in a steadily growing company. The percentage of technology spending in relation to revenue will show a steady decline year over year, as illustrated in Figure 3.1.

Figure 3.1 Neglected IT. Flat IT spending in a growing company.

IT Distribution Analysis

IT distribution analysis is a technique used to understand the relative amounts of people, spending, applications, customers, etc. in relation to the amount of technology maintained and developed in an organization.

This data reveals the current level of resource utilization which achieves the current output or level of support. You can think of this as the miles per gallon (or kilometers per liter) rating for your technology investment. The results are points for understanding and discussion within the current organization. There is also industry data available if a comparison to peer groups is desired (see Figure 4.1 through Figure 4.3 in Chapter 4).

It can be very productive to analyze the cost of maintenance as a percentage of the overall IT budget. Jim Fister of Intel notes that "as much as 90 percent of today's IT budgets go to maintenance." If maintenance expenses are 80 percent or more of the IT budget, then there should be room to make savings to devote more funds to new capabilities. Follow-up questions address which of the applications require the most support and what solutions are more effective.

The same approach can be applied to each technology category. The analysis should be performed according to categories that already represent the way your IT Providers support the client base, usually by departments of sales, marketing, operations, human resources, finance, manufacturing, etc. (see Figure 3.2). Another approach is analysis by core process area. This can be more difficult but it provides more insight (see Figure 3.3).

Figure 3.2 IT spending by supported client base (example).

Figure 3.3 IT spending by function.

The next question to ask is whether the IT budget is proportionally targeted to the right areas to achieve the business strategy.

Looking at the percentage distribution for each department as a pie chart (see Figure 3.2), can the portions be reallocated to solve business problems or client dissatisfaction or better align with business strategy? For example, if 30 percent of the company's non-IT budget is spent on marketing this year, then should 30 percent of the IT budget (or the "discretionary" IT budget) also be allocated to marketing?

As an aside, the "discretionary" IT budget is one name for that part of the IT budget that is available to be spent each year "at the company's discretion"; that is, IT budget that is not already committed to maintenance. As noted previously, this is often a minority of the budget.

The results of this internal analysis can be illuminating. The discussions often generate champions within the client base for increased funding for IT. Other observations of organizational need and pain points surface which can form grounds for conversations about change.

An important point for discussion, though not obvious, is "How did we get here?" What elements are present in the culture or structure of the organization that led to the current state?

What organizational factors, cultural norms, or customer demands contributed to the current state of spending or distribution of IT resources? This determines the influencing factors such as operations receiving the greatest share of IT resources (see Figure 3.3).

What is the prioritization process for approving new IT applications or enhancements? Failing to prioritize often leads to spending on the wrong things or the squeaky wheel project (or people).

Does a target architecture exist? Are design standards adhered to? Conversely, persisting with antiquated disparate systems significantly increases maintenance costs and allows for many costly independent acquisition or development paths to emerge. Having a vision of the target solution allows measurement of progress toward the target and a sense of what "done" looks like.

Watts Humphrey once said, "Watch what people do when something goes wrong, that's the process." What is the process in the IT organization? What do people do under stress? Avoiding key steps in a standard process erodes quality and contributes to a chaotic culture where deviations become the way of doing business.

Answering the questions presented here is critical to understanding the current state of performance for the organization. The answers are also essential to understanding how the current state of spending came to be. To make lasting changes, the norms of behavior must also change. An excellent source for more on this topic is Leading Change by John Kotter.2

Organizational Evaluation

Organizational evaluation is an evaluation of the skills needed to run the IT organization. The human assets of the organization are always a delicate topic; it is important to take every step in this process or none at all.

1. Establish guiding principles. These principles become the guardrails for the evaluation. The guiding principles should answer the question: "Why are we doing this?" Some examples are

  • Align IT outlets to deliver business strategy
  • Enable our target architecture
  • Evaluate IT spending
  • Support the need to consolidate
  • Support the need for rapid growth
  • To fund a new function
  • We're taking a disciplined approach to realigning the organization

The principles you establish are used to confirm the resulting actions, ensuring alignment with the original intent of the organizational evaluation.

2. Engage the clients. Often called "users," the clients of the IT Providers have plenty of valuable input to determine the skills lacking in IT. This engagement can be captured through survey data, designed workshops, and focus groups. Prioritize the results. Remove outliers, observations that might be compelling but very rare. Seek to solve for the top three to five items.

Client engagement is necessary to understand what issues really need to be solved. A lack of role clarity or no clear support might influence moving to a client-focused allocation of IT resources. This structure would align whole teams to focus on a specific business area. This structure is excellent for team focus, aligned goals, and client satisfaction. The trade-off is a poor structure for functional support (all developers are not in the same group). It often costs more and can be difficult for central roles like architecture to support so many groups. As such, it is an important decision point in IT budget allocations.

3. Engage IT Providers. Many people within the IT organization from bottom to top possess knowledge about what needs to be corrected. They will also have valuable information about the existence of barriers to change and the associated risks. Again, prioritize the results. You should see overlap with the theme of the input from the clients.

4. Establish goals for the outcome of this exercise. These goals will be reviewed with the leadership in the organization, at a minimum. Include the time frame in which the results will be announced, the sooner the better. These goals could include onshore to offshore ratios of staff, like a 60/40 or 70/30 split, improving the client perception of IT, or faster throughput. Regardless of the goals of the exercise, formal documentation is required.

5. Skills evaluation. Start with the future state of the organization. Use documents like the three-year business strategy, technology target architecture, or other forward-looking information to define high-level position types. Use a table similar to this to analyze the results.

Future RoleFuture CompetenciesFuture Technical SkillsNumber RequiredCan This Position be Outsourced (why or why not)
Business analyst Facilitation skills
Requirements elicitation
Use case development
Risk analyis
Project manager Communication
Earned value management
Business case development
10 Space
Development manager Space Space 4 Space
Process engineer Space CMMI
3 Space>
Architect Space> Business and technology knowledge 2 Space
Developer Documentation List technologies needed 30 Space
Tester Space List testing tools 15 Space

6. Select the organization design. In technology there are two clear choices or a hybrid approach:

  • A functional alignment of groups and reports of like functions, (e.g., analysts, project managers, developers, testers, etc.). This design may be the most efficient because like functions are in a single team. Its limitations can be the lack of alignment with client goals, and IT roles and teams that are siloed.
  • A client-focused design aligns entire delivery units according to client groups. This often entails diverse roles reporting to one manager (e.g., project mangers, analysts, developers, and testers). Benefits include high client satisfaction and faster delivery through greater focus. Limitations of this design include less efficiency through teams developing their own processes, duplicating functions, and not providing clear priorities for shared resources like design or architecture.

7. Assess the organizational design according to the goals from step 4. Ensure the new design will fill the gaps identified. Communicate the results to leadership, plan for the communication and rollout of the changes.

Containing Cost Versus Innovation

Striking the right balance between innovation and controlling IT cost is a continual challenge. Being innovative means some initiatives will fail. To breed a culture where risks are taken, failure must be an acceptable, even celebrated, result. It took Thomas Edison thousands of experiments to find the right amount of the right material to make a long-lasting light bulb. Edison viewed each failure as one more option eliminated, thereby getting closer to his goal. Building for innovation and developing a culture for taking risk does not need to clash with cost consciousness. However, an appropriate structure or system of support to foster innovation is necessary.

The following steps establish innovation as an important element of the IT culture and budget:

  • Establish a budget for innovation.
  • Assign responsibility for innovation to a senior member of the IT organization.
  • Set annual goals for innovation.
  • Provide the time necessary for staff to work on innovative ideas.
  • Celebrate failure.


This chapter describes a general philosophy and detailed measures for starting a discussion about whether "enough" is being spent by the business on its IT.


1. Gartner Consulting Worldwide IT Benchmark Service Trends and Findings for 2007.
2. Kotter, John. 1996. Leading Change. Harvard Business School Press.

About the Author

From The Business Value of IT: Managing Risks, Optimizing Performance and Measuring Results by Michael D. S. Harris, David Herron and Stasia Iwanicki. New York: Auerbach Publications, 2008.
Free Subscription to IT Today

Powered by VerticalResponse


Guided Insights

Guided Insights helps global project teams speed time to results through better collaboration across time zones, cultures and other boundaries. Special areas of focus are remote team leadership, facilitation skills, virtual team collaboration, project jumpstart workshops and design and facilitation of virtual meetings.

Gantt Chart software for project planning and scheduling

© Copyright 2008 Auerbach Publications