Total cost of ownership (TCO) studies of software—particularly OS software—generate a lot of splash in the trade press. But the reality is that few people look carefully at TCO studies and the thinking that goes into them, and fewer still have access to the in-depth data that enters into the formulation of the studies.

Recently, two major TCO studies about Windows 2000 offer nuanced looks at Win2K with almost no solid conclusions—just a lot of conditionals. According to the press, GartnerGroup’s "Windows 2000 Migration—Finding ROI" says that Win2K is too expensive, and Giga Group’s "Windows 2000 Benefits Are Key to Upgrade Decision" says that Win2K is worth the money. We found that most companies don’t fit the assumptions of the reports, and that the analysts intend their reports to serve as models only, meant to guide an enterprise in calculating its migration and deployment costs and benefits.

What Is TCO Analysis?
TCO analysis exposes a product's hidden costs over its projected life cycle. Win2K's TCO isn’t just the software's cost—it’s the cost of hardware upgrades a company must make to ensure that its systems are Win2K-compatible, the cost of training IT staff on Win2K, the cost of labor for installing the OS, and the cost of productivity as users get used to the new product.

Figure 1 shows GartnerGroup's cost breakdown for a worst-case scenario for migrating from Windows 9x to Windows 2000 Professional (Win2K Pro). As you can see, Win2K's licenses are a small proportion of the total.

The first thing to understand about TCO is that you can't calculate it in a vacuum. A product's TCO isn't fixed or static; it changes from company to company. Does a company have top-of-the-line hardware or outdated systems? TCO is higher for a cutting-edge product if a company is trying to install it on older systems. What OS is the company migrating from? Migrating from Novell NetWare to Win2K, for example, costs more than migrating from Windows NT 4.0 to Win2K.

Methodologies
The complexity of calculating TCO results in widely varying methodologies. GartnerGroup, for example, uses a Stalking Horse model, which means that it calculates TCO for an imaginary company with particular qualities. In the model that GartnerGroup used for its Win2K TCO study, the imaginary company employed 2500 users, used no laptops, and didn't require that people travel between sites during the installation process. The Stalking Horse concept is critical to understanding GartnerGroup’s numbers, but the press rarely reported it. GartnerGroup attaches this disclaimer to its migration study: "Stalking Horses do not necessarily reflect the reality of any particular IS organization. We encourage enterprises to perform their own analyses using our own models."

Giga Group’s analysis, called the Total Economic Impact (TEI), reflects its focus on balancing Win2K's total cost with its overall benefit to a company. The benefits component, which represents a more complete analysis of the overall impact of moving to Win2K, reflects Giga Group's tendency to speak to business managers.

No discernable hard-and-fast methodology exists for how GartnerGroup or Giga Group collected the data in their reports. Unlike some other research analyst firms, neither company took a survey-and-average approach. Instead, analysts at each firm told us that they used a combination of interviews with people in the industry, personal industry experience, and simple poking around. As we'll see, this approach can lead to incredible variations in conclusions.

Is Pro the Win in Win2K?
GartnerGroup and Giga Group differ considerably in their TCO estimates for migrating to Win2K Pro. GartnerGroup estimates that migrating from NT Workstation to Win2K Pro will cost between $1268 and $2044 per desktop, and estimates that migration from Win9x to Win2K Pro will cost an astounding $2015 to $3191 per system. Giga Group, on the other hand, estimates a per-system cost of $973 for a similar migration. Even GartnerGroup’s migration cost estimates for best-case scenarios are higher than Giga Group’s single estimate.

Why the discrepancy? GartnerGroup considers several factors that pertain to upgrading to Win2K Pro that Giga Group ignores. GartnerGroup adds in the cost for testing Win2K Pro compatibility with custom applications built for an older OS, along with the development costs for modifying said applications. GartnerGroup, which believes that 15 to 20 percent of applications will be incompatible with Win2K Pro and will need testing and replacing, estimates application testing and development costs at $23 to $287 per migration for its model company. Giga Group, however, thinks that such costs are marginal when amortized across all desktops in an enterprise.

Giga estimates that the total worst-case per-system labor will amount to about 2 hours; the company estimates this cost at $140 per system. GartnerGroup offers a more intricate model that takes into account fixed labor costs such as project planning, contract negotiations, and Help desk training. Fixed costs are large-scale project costs that are relatively independent of the particular number of desktops. For migration from Win9x, GartnerGroup estimates that fixed costs break down to between $52 and $165 per desktop—add to this cost the company's estimates for system-by-system installation. For Win9x migration, GartnerGroup estimates between 11.75 and 42.5 hours of labor per system, which works out to between $277 and $1018 in labor per desktop. So, GartnerGroup’s worst-case-scenario estimate is seven times greater than Giga Group’s worst-case estimate for labor costs.

How did GartnerGroup arrive at these figures? Let’s examine the company's worst-case scenario. It estimates 1 hour for installation preparation, 6 hours for adding new hardware to each desktop, 6.5 hours for upgrading hardware, 3 hours for installing the software, 4 hours for restoring and testing the system, 8 hours for user training, and 14 hours for the user to get used to Win2K Pro and restore the system. As Figure 2 shows, these estimates add up to over 5 days of labor for migrating each desktop to Win2K Pro.

The differences between GartnerGroup's and Giga Group's per-system cost analysis pale when compared to the differences these analysts measure in their benefits analysis. Giga Group tries to figure out the economic benefit of a Win2K Pro migration by estimating that the new OS will add 3 to 5 percent to the productivity of each employee. For an 8-hour day, 3 to 5 percent is about 10 to 24 minutes. According to Rob Enderle, a Giga Group vice president and a report author, "The majority of these savings come from Win2K Pro’s increased reliability. \[An older Windows desktop\] requires a reboot once every 24 hours. When an application goes down, it takes down the whole environment. You can lose everything you’ve done since the last save. This doesn’t happen with Win2K Pro \[as often\]."

Using these estimates and assuming that the average employee cost is $100,000 a year, Win2K Pro adds $3000 to $5000 of labor value per user to an employer's bottom line. Giga Group notes in its report that, on average, "companies generally make over five times labor cost in revenue." If you accept this principle, migrating to Win2K Pro would result in a revenue increase of $15,000 to $25,000, and the migration would pay for itself in months.

Unlike the Giga Group, GartnerGroup makes no benefit calculations in its report. In its central study, "Windows 2000: Desktop Migration Costs," GartnerGroup concludes that Win2K Pro will offer "little return on investment (ROI) over the useful system life," but it offers no benefit analysis. In interviews with GartnerGroup's Win2K analyst Mike Silver, and in examining some early drafts of unpublished GartnerGroup documents, we discovered the research firm's rationale for its limited view of Win2K Pro benefits. GartnerGroup notes that Win2K Pro offers few hard benefits (i.e., measurable benefits that an IS staff can include in its budget, such as discrete, quantifiable savings in staff labor). Most savings are in soft costs, such as savings in computer downtime and productivity spread across all computer users in a company. Because these savings are soft, Gartner ignores them.

By ignoring Win2K Pro's benefits, GartnerGroup has clearly aimed its report at an IS staff planner trying to make decisions about migration that will look best on an IS staff budget rather than an enterprise leader trying to maximize productivity for his or her company.

GartnerGroup and Giga Group each provide ROI estimates for migrating from Win9x to Win2K Pro. GartnerGroup’s conclusion of a poor ROI (i.e., $0) for Win2K Pro is based on its proposition of zero-quantifiable-ROI. Giga Group's ROI far outweighs even GartnerGroup’s worst-case TCO estimate.

What if we were to assume an extra-conservative approach to Giga Group’s estimates by taking its worst-case estimates and cutting them in half? Giga Group estimates that the productivity increase that will result from Win2K Pro is 3 to 5 percent; let’s assume that it’s 1.5 percent. Giga Group estimates that companies make five times labor costs; let’s halve that to 2.5 times. With our conservative estimates, and assuming that the average employee cost is $100,000, we arrive at the following simple calculation for Win2K ROI:

Revenue = $100,000 salary x 1.5 percent productivity increase x 2.5 labor = $3750 gain per user per year.

Let’s go further. If we look at Gartner Group’s worst-case scenario for migrating from Win9x, which estimates 8640 hours in application development and 14 hours per desktop in end-user acclimatization, the per-desktop TCO for Win2K Pro is $3191—less than our extra-conservative ROI of $3750.

So, if we halve all the figures in Giga Group's benefits estimate and pit those figures against Gartner Group's very worst-case scenario, Win2K Pro still more than pays for itself in less than a year. (If we use Gartner Group's best-case scenario for migrating from NT Workstation to Win2K Pro, the new OS pays for itself in 4 months. If Giga Group's assumptions hold true for your company, migrating to Win2K Pro should be an easy decision.

GartnerGroup’s Win2K Pro TCO model is more detailed, and has value if you're calculating your TCO analysis. While some of its worst-case scenario estimates seem extreme, even as worst-case estimates, the detail and all-encompassing nature of its calculations is edifying. For calculating the total cost of the operation, most corporate analysts will find GartnerGroup’s model useful. However, when GartnerGroup generated its numbers, the company made several assumptions about a typical organization, and so GartnerGroup analysts tell their clients to run the numbers for themselves.

That said, GartnerGroup’s ROI analysis falls short. Without measuring the benefits in a cost-benefit analysis, you can’t accurately measure the impact of Win2K Pro adoption on an entire organization. Such benefits aren't important when you're simply considering the impact of Win2K Pro on an IT budget, but just because the benefits are hard to quantify doesn’t mean they don't exist. And as our calculations indicate, even a conservative estimate of the benefits shows that Win2K Pro pays for itself in under a year.

Windows 2000 Server: Just Like Starting Over
Calculating Windows 2000 Server (Win2K) Server TCO is much harder than calculating Win2K Pro TCO, and calculating benefits is nearly impossible. For Win2K Server, GartnerGroup and Giga Group again set up model companies— GartnerGroup stayed with its 2500-person company, and Giga Group used an imaginary 5000-person company. Trying to summarize the studies' Win2K Server migration costs is exceedingly difficult. GartnerGroup, for example, presents entirely different estimates for migrating to Win2K Server from different server OSs. Its estimates range from $279 per user for migration from NT Server 4.0 to $457 per user for migration from NetWare 3.x.

How do GartnerGroup's and Giga Group's migration costs compare? GartnerGroup’s NT 4.0-to-Win2K Server migration cost estimate is $279 per user; Giga Group’s estimate is $107 per user. Much of the difference arises from radically different initial assumptions. Giga Group based its calculations on 30 servers for 5000 users—167 users per server; GartnerGroup assumed 33 servers for 2500 users—76 users per server. This difference highlights the fact that in TCO calculations for server OSs, sweeping one-size-fits-all conclusions don't cut it. Depending on the business, different servers run different applications and support different numbers of clients. Change the mix, and the number of clients that a server can support changes dramatically.

When you step away from the numbers themselves, the overall advice these analysts give sounds pretty much the same. GartnerGroup and Giga Group advise clients to wait 6 to 9 months after Win2K Server's release to migrate. And, each company tells its clients, in almost exactly the same words, that they should have a sound business reason to migrate and should calculate their own ROI.

Still, Giga Group sounds more optimistic about the possibility of a positive ROI for Win2K Server than does GartnerGroup. The key arguments in the server analyses revolve around the viability of policy management and the success of adopting third-party tools.

Win2K Server’s policy management features present a critical ease-of-use factor, and thus a labor-savings factor, for this new OS. Giga Group predicts that policy management features will result in tremendous savings—although its study doesn’t provide any numerical analysis.

According to GartnerGroup's Mike Silver, if a company manages to get Win2K Server's policy management features up and running, the ROI would indeed be strongly positive. However, he cautioned that simply buying a server package that includes policy management features doesn't easily translate into integrating policy management features into the workplace. Silver pointed out that many policy management products are already available. "If you failed to put a policy-management product into place earlier, there’s probably a reason for it, and that reason will keep you from getting the most out of Win2K Server’s policy management features," he said.

Another area of analyst contention concerns the migration of NT 4.0 domain structures to Active Directory (AD). Giga Group’s analysis discusses third-party AD management tools, such as Mission Critical Software's, Entevo's, and FastLane's, at length and concludes that they add "business-crucial event, application, and policy management functionality not available in Active Directory" \[and\] "provides customers with instant and sustainable ROI." Giga Group concludes that "this type of functionality will be a ‘must buy’ for any enterprise organization that plans to make extensive use of AD ... The resulting savings in administrative time will, to a large extent, mitigate the capital expenditure and help firms realize ROI within the first 6 months of installation."

Alternatively, GartnerGroup completely ignores third-party management tools. Its analysis only pauses to note, "Our base migration cost model does not include use of tools or best practices, which can help lower migration costs."

\[Barrie: Can we close with a statement about who's better off following GartnerGroup's server recommendations vs. Giga Group's server recommendations? Rob\]

Total cost of ownership (TCO) studies of software—particularly OS software—generate a lot of splash in the trade press. But the reality is that few people look carefully at TCO studies and the thinking that goes into them, and fewer still have access to the in-depth data that enters into the formulation of the studies.

Recently, two major TCO studies about Windows 2000 offer nuanced looks at Win2K with almost no solid conclusions—just a lot of conditionals. According to the press, GartnerGroup’s "Windows 2000 Migration—Finding ROI" says that Win2K is too expensive, and Giga Group’s "Windows 2000 Benefits Are Key to Upgrade Decision" says that Win2K is worth the money. We found that most companies don’t fit the assumptions of the reports, and that the analysts intend their reports to serve as models only, meant to guide an enterprise in calculating its migration and deployment costs and benefits.

What Is TCO Analysis?
TCO analysis exposes a product's hidden costs over its projected life cycle. Win2K's TCO isn’t just the software's cost—it’s the cost of hardware upgrades a company must make to ensure that its systems are Win2K-compatible, the cost of training IT staff on Win2K, the cost of labor for installing the OS, and the cost of productivity as users get used to the new product.

Figure 1 shows GartnerGroup's cost breakdown for a worst-case scenario for migrating from Windows 9x to Windows 2000 Professional (Win2K Pro). As you can see, Win2K's licenses are a small proportion of the total.

The first thing to understand about TCO is that you can't calculate it in a vacuum. A product's TCO isn't fixed or static; it changes from company to company. Does a company have top-of-the-line hardware or outdated systems? TCO is higher for a cutting-edge product if a company is trying to install it on older systems. What OS is the company migrating from? Migrating from Novell NetWare to Win2K, for example, costs more than migrating from Windows NT 4.0 to Win2K.

Methodologies
The complexity of calculating TCO results in widely varying methodologies. GartnerGroup, for example, uses a Stalking Horse model, which means that it calculates TCO for an imaginary company with particular qualities. In the model that GartnerGroup used for its Win2K TCO study, the imaginary company employed 2500 users, used no laptops, and didn't require that people travel between sites during the installation process. The Stalking Horse concept is critical to understanding GartnerGroup’s numbers, but the press rarely reported it. GartnerGroup attaches this disclaimer to its migration study: "Stalking Horses do not necessarily reflect the reality of any particular IS organization. We encourage enterprises to perform their own analyses using our own models."

Giga Group’s analysis, called the Total Economic Impact (TEI), reflects its focus on balancing Win2K's total cost with its overall benefit to a company. The benefits component, which represents a more complete analysis of the overall impact of moving to Win2K, reflects Giga Group's tendency to speak to business managers.

No discernable hard-and-fast methodology exists for how GartnerGroup or Giga Group collected the data in their reports. Unlike some other research analyst firms, neither company took a survey-and-average approach. Instead, analysts at each firm told us that they used a combination of interviews with people in the industry, personal industry experience, and simple poking around. As we'll see, this approach can lead to incredible variations in conclusions.

Is Pro the Win in Win2K?
GartnerGroup and Giga Group differ considerably in their TCO estimates for migrating to Win2K Pro. GartnerGroup estimates that migrating from NT Workstation to Win2K Pro will cost between $1268 and $2044 per desktop, and estimates that migration from Win9x to Win2K Pro will cost an astounding $2015 to $3191 per system. Giga Group, on the other hand, estimates a per-system cost of $973 for a similar migration. Even GartnerGroup’s migration cost estimates for best-case scenarios are higher than Giga Group’s single estimate.

Why the discrepancy? GartnerGroup considers several factors that pertain to upgrading to Win2K Pro that Giga Group ignores. GartnerGroup adds in the cost for testing Win2K Pro compatibility with custom applications built for an older OS, along with the development costs for modifying said applications. GartnerGroup, which believes that 15 to 20 percent of applications will be incompatible with Win2K Pro and will need testing and replacing, estimates application testing and development costs at $23 to $287 per migration for its model company. Giga Group, however, thinks that such costs are marginal when amortized across all desktops in an enterprise.

Giga estimates that the total worst-case per-system labor will amount to about 2 hours; the company estimates this cost at $140 per system. GartnerGroup offers a more intricate model that takes into account fixed labor costs such as project planning, contract negotiations, and Help desk training. Fixed costs are large-scale project costs that are relatively independent of the particular number of desktops. For migration from Win9x, GartnerGroup estimates that fixed costs break down to between $52 and $165 per desktop—add to this cost the company's estimates for system-by-system installation. For Win9x migration, GartnerGroup estimates between 11.75 and 42.5 hours of labor per system, which works out to between $277 and $1018 in labor per desktop. So, GartnerGroup’s worst-case-scenario estimate is seven times greater than Giga Group’s worst-case estimate for labor costs.

How did GartnerGroup arrive at these figures? Let’s examine the company's worst-case scenario. It estimates 1 hour for installation preparation, 6 hours for adding new hardware to each desktop, 6.5 hours for upgrading hardware, 3 hours for installing the software, 4 hours for restoring and testing the system, 8 hours for user training, and 14 hours for the user to get used to Win2K Pro and restore the system. As Figure 2 shows, these estimates add up to over 5 days of labor for migrating each desktop to Win2K Pro.

The differences between GartnerGroup's and Giga Group's per-system cost analysis pale when compared to the differences these analysts measure in their benefits analysis. Giga Group tries to figure out the economic benefit of a Win2K Pro migration by estimating that the new OS will add 3 to 5 percent to the productivity of each employee. For an 8-hour day, 3 to 5 percent is about 10 to 24 minutes. According to Rob Enderle, a Giga Group vice president and a report author, "The majority of these savings come from Win2K Pro’s increased reliability. \[An older Windows desktop\] requires a reboot once every 24 hours. When an application goes down, it takes down the whole environment. You can lose everything you’ve done since the last save. This doesn’t happen with Win2K Pro \[as often\]."

Using these estimates and assuming that the average employee cost is $100,000 a year, Win2K Pro adds $3000 to $5000 of labor value per user to an employer's bottom line. Giga Group notes in its report that, on average, "companies generally make over five times labor cost in revenue." If you accept this principle, migrating to Win2K Pro would result in a revenue increase of $15,000 to $25,000, and the migration would pay for itself in months.

Unlike the Giga Group, GartnerGroup makes no benefit calculations in its report. In its central study, "Windows 2000: Desktop Migration Costs," GartnerGroup concludes that Win2K Pro will offer "little return on investment (ROI) over the useful system life," but it offers no benefit analysis. In interviews with GartnerGroup's Win2K analyst Mike Silver, and in examining some early drafts of unpublished GartnerGroup documents, we discovered the research firm's rationale for its limited view of Win2K Pro benefits. GartnerGroup notes that Win2K Pro offers few hard benefits (i.e., measurable benefits that an IS staff can include in its budget, such as discrete, quantifiable savings in staff labor). Most savings are in soft costs, such as savings in computer downtime and productivity spread across all computer users in a company. Because these savings are soft, Gartner ignores them.

By ignoring Win2K Pro's benefits, GartnerGroup has clearly aimed its report at an IS staff planner trying to make decisions about migration that will look best on an IS staff budget rather than an enterprise leader trying to maximize productivity for his or her company.

GartnerGroup and Giga Group each provide ROI estimates for migrating from Win9x to Win2K Pro. GartnerGroup’s conclusion of a poor ROI (i.e., $0) for Win2K Pro is based on its proposition of zero-quantifiable-ROI. Giga Group's ROI far outweighs even GartnerGroup’s worst-case TCO estimate.

What if we were to assume an extra-conservative approach to Giga Group’s estimates by taking its worst-case estimates and cutting them in half? Giga Group estimates that the productivity increase that will result from Win2K Pro is 3 to 5 percent; let’s assume that it’s 1.5 percent. Giga Group estimates that companies make five times labor costs; let’s halve that to 2.5 times. With our conservative estimates, and assuming that the average employee cost is $100,000, we arrive at the following simple calculation for Win2K ROI:

Revenue = $100,000 salary x 1.5 percent productivity increase x 2.5 labor = $3750 gain per user per year.

Let’s go further. If we look at Gartner Group’s worst-case scenario for migrating from Win9x, which estimates 8640 hours in application development and 14 hours per desktop in end-user acclimatization, the per-desktop TCO for Win2K Pro is $3191—less than our extra-conservative ROI of $3750.

So, if we halve all the figures in Giga Group's benefits estimate and pit those figures against Gartner Group's very worst-case scenario, Win2K Pro still more than pays for itself in less than a year. (If we use Gartner Group's best-case scenario for migrating from NT Workstation to Win2K Pro, the new OS pays for itself in 4 months. If Giga Group's assumptions hold true for your company, migrating to Win2K Pro should be an easy decision.

GartnerGroup’s Win2K Pro TCO model is more detailed, and has value if you're calculating your TCO analysis. While some of its worst-case scenario estimates seem extreme, even as worst-case estimates, the detail and all-encompassing nature of its calculations is edifying. For calculating the total cost of the operation, most corporate analysts will find GartnerGroup’s model useful. However, when GartnerGroup generated its numbers, the company made several assumptions about a typical organization, and so GartnerGroup analysts tell their clients to run the numbers for themselves.

That said, GartnerGroup’s ROI analysis falls short. Without measuring the benefits in a cost-benefit analysis, you can’t accurately measure the impact of Win2K Pro adoption on an entire organization. Such benefits aren't important when you're simply considering the impact of Win2K Pro on an IT budget, but just because the benefits are hard to quantify doesn’t mean they don't exist. And as our calculations indicate, even a conservative estimate of the benefits shows that Win2K Pro pays for itself in under a year.

Windows 2000 Server: Just Like Starting Over
Calculating Windows 2000 Server (Win2K) Server TCO is much harder than calculating Win2K Pro TCO, and calculating benefits is nearly impossible. For Win2K Server, GartnerGroup and Giga Group again set up model companies— GartnerGroup stayed with its 2500-person company, and Giga Group used an imaginary 5000-person company. Trying to summarize the studies' Win2K Server migration costs is exceedingly difficult. GartnerGroup, for example, presents entirely different estimates for migrating to Win2K Server from different server OSs. Its estimates range from $279 per user for migration from NT Server 4.0 to $457 per user for migration from NetWare 3.x.

How do GartnerGroup's and Giga Group's migration costs compare? GartnerGroup’s NT 4.0-to-Win2K Server migration cost estimate is $279 per user; Giga Group’s estimate is $107 per user. Much of the difference arises from radically different initial assumptions. Giga Group based its calculations on 30 servers for 5000 users—167 users per server; GartnerGroup assumed 33 servers for 2500 users—76 users per server. This difference highlights the fact that in TCO calculations for server OSs, sweeping one-size-fits-all conclusions don't cut it. Depending on the business, different servers run different applications and support different numbers of clients. Change the mix, and the number of clients that a server can support changes dramatically.

When you step away from the numbers themselves, the overall advice these analysts give sounds pretty much the same. GartnerGroup and Giga Group advise clients to wait 6 to 9 months after Win2K Server's release to migrate. And, each company tells its clients, in almost exactly the same words, that they should have a sound business reason to migrate and should calculate their own ROI.

Still, Giga Group sounds more optimistic about the possibility of a positive ROI for Win2K Server than does GartnerGroup. The key arguments in the server analyses revolve around the viability of policy management and the success of adopting third-party tools.

Win2K Server’s policy management features present a critical ease-of-use factor, and thus a labor-savings factor, for this new OS. Giga Group predicts that policy management features will result in tremendous savings—although its study doesn’t provide any numerical analysis.

According to GartnerGroup's Mike Silver, if a company manages to get Win2K Server's policy management features up and running, the ROI would indeed be strongly positive. However, he cautioned that simply buying a server package that includes policy management features doesn't easily translate into integrating policy management features into the workplace. Silver pointed out that many policy management products are already available. "If you failed to put a policy-management product into place earlier, there’s probably a reason for it, and that reason will keep you from getting the most out of Win2K Server’s policy management features," he said.

Another area of analyst contention concerns the migration of NT 4.0 domain structures to Active Directory (AD). Giga Group’s analysis discusses third-party AD management tools, such as Mission Critical Software's, Entevo's, and FastLane's, at length and concludes that they add "business-crucial event, application, and policy management functionality not available in Active Directory" \[and\] "provides customers with instant and sustainable ROI." Giga Group concludes that "this type of functionality will be a ‘must buy’ for any enterprise organization that plans to make extensive use of AD ... The resulting savings in administrative time will, to a large extent, mitigate the capital expenditure and help firms realize ROI within the first 6 months of installation."

Alternatively, GartnerGroup completely ignores third-party management tools. Its analysis only pauses to note, "Our base migration cost model does not include use of tools or best practices, which can help lower migration costs."