Notebook Total Cost of Ownership: 2013 Update

24 May 2013 ID:G00250520
Analyst(s): Federica Troni

VIEW SUMMARY

The gap between desktop and notebook total cost of ownership continues to close, but tablets are also being considered as notebook replacements. Infrastructure and operations leaders should use this analysis to support their endpoint adoption decision.

Overview

Key Challenges

  • Organizations struggle to recognize that there are different categories of notebook users and that each user group requires a different level of support.
  • As notebook prices continue to decline, their total cost of ownership (TCO) remains higher than that of a desktop, forcing organizations to focus on the cost-benefit of replacing a desktop with a notebook.
  • Organizations need to look into their users' requirements to determine when tablets can be used as companion devices or notebook replacements, and they need to investigate the TCO implications.

Recommendations

  • Segment your notebook users to understand their working patterns and profiles.
  • Equip users with notebooks when mobility and offline capability allow them to increase productivity.
  • Use management tools, and develop skills and processes to reduce notebook TCO, but recognize that full lockdown (removing administrator rights) remains inappropriate for most traveling notebook workers.
  • Examine bring your own device (BYOD) programs for tablets that are companion devices, if your corporate environment allows, or revise lightweight notebook standards to include hybrid or convertible devices as replacements, and recognize that tablets generally raise the overall TCO per workspace.
  • Consider media tablets as notebook replacements to reduce TCO for users who need minimal or no access to Windows applications, and Windows 8 tablets for users who need more access to Windows applications, but ensure everyone understands the trade-offs.

Table of Contents

Introduction

This document was revised on 22 August 2013. The document you are viewing is the corrected version. For more information, see the Corrections page on gartner.com.

The main purpose of our client computing TCO research is to help IT leaders, CIOs and CFOs understand the changes in the cost of a client computing deployment that may occur when making certain changes, such as:

  • Best practices — increase the level of security and management
  • Platforms — impact of substitution of desktops versus notebooks versus mobile devices
  • Usage patterns — the level of user mobility
  • Architecture — traditional PCs, server-based computing (SBC), hosted virtual desktops (HVDs), etc.
  • Length of asset ownership — one to six years
  • Number of applications — an environment that uses more applications will be more expensive to manage

This specific analysis focuses on notebooks.

Analysis

Recognize What's New in Notebook TCO in 2013

As with all of Gartner's TCO models, organizations should use our framework with their own numbers, where possible.

Hardware, Software and Facilities Costs

We determine the cost of notebooks using the list prices of business-class PCs that match our recommended configurations for mainstream users (see "Recommended Configurations for Enterprise Notebooks, 2013").

We are including $1,045 for a current mainstream business-class notebook configuration, with a life cycle of three years, a 5% decline since our last update. As usual, we include notebook list prices, without volume discounts. Organizations that want to calculate their own notebook TCO should use the effective price paid for the device, including volume discounts and the correct life cycle.

We have not made significant changes to our cost for end-user software ($632) in this iteration of the research, except for the addition of a small amount ($7) to represent a cost allocation for e-learning modules. For well-managed scenarios, we have reduced software costs (and only for their maintenance by 3%) to reflect an uptake of software usage monitoring tools.

IT Operations and Administration Costs

We increased the IT labor costs by 5% in the base salaries we use for our scenarios (see Table 1). These salaries are typical of the high costs in mature markets, and should reflect the reality of North American and Western European countries. The model also accounts for a 33% burden rate. Organizations, especially those operating in markets with substantially lower salaries and/or very different burden rates, should replace our sample numbers with their own estimates.

Table 1. Salaries for IT Personnel in Mature Markets, Unburdened

IT Personnel

Salaries

Tier 1

$51,040

Tier 2

$72,345

Tier 3

$94,605

Manager

$89,040

Source: Gartner (May 2013)

While IT labor costs have increased slightly, we have modified some elements that decrease the complexity and cost of the environment.

We assume that the number of Web-based, thin, hosted, OS-neutral applications have reached 55% of the typical application mix of organizations (see "Windows Applications Will Be Critical Through the Planning Horizon, but Lose Majority in 2012"). A higher percentage of Web-based and hosted applications decreases the effort required to manage user images, which is reflected in our numbers. We have forecast that 20% of applications will be virtualized, reducing the time spent on regression testing and minimizing application conflict issues.

Following feedback received from our clients and based on evidence from our primary research, we made some adjustments to our assumptions around the ratio of full-time equivalents (FTEs) to users. The FTE requirements for Tier 1 support have increased by 11%, and the FTE ratio for Tier 2 and Tier 3 declined by 20%. Administration and management costs have decreased by a similar percentage.

Training for end users and IT personnel is no longer provided solely through classes, but tends to be a blend of e-learning modules, live instruction and internal mentorship. Therefore, we have decreased our allocation slightly for organizing traditional classes, and have added a yearly cost allocation in the software category to account for the purchase of e-learning modules.

End-User Costs

In line with our findings, we have revised end-user salaries, increasing them by 5% (see Table 2). The end-user salaries shown in our TCO research are used to estimate the cost of lost productivity due to planned and unplanned downtime, formal and informal training, and users' attempts to resolve their own IT issues or those of their peers, instead of or before using formal IT support channels. Since salaries significantly affect all end-user costs, organizations should change these costs as appropriate.

Table 2. Salaries for End Users in Mature Markets, Unburdened

End Users

Salaries

Data Entry

$27,825

Task-Oriented

$50,085

Knowledge Worker

$89,040

Power User

$166,950

Source: Gartner (May 2013)

Levels of Manageability, OS Versions and Applications

As in previous notebook TCO research, we model a deployment of 2,500 desktop users, all in one location, with a centralized IT organization. We have created four scenarios, which assume that different levels of manageability are applied:

  • Unmanaged — Users can install applications and change settings; few to no management tools are being used.
  • Somewhat managed — Some management tools are implemented, but processes and policies are not fully developed.
  • Moderately managed — Tools and good processes and policies are in place; users can install software and change some settings.
  • Locked and well-managed — There are tools, processes and policies; users cannot install software or change critical settings.

Table 3 summarizes the user mix in these scenarios. There are no changes in the notebook user mix from 2011.

Table 3. Notebook TCO 2013, User Mix

Type of User

Day Extenders

Traveling Workers

Data Entry Workers

0%

0%

Task-Oriented Workers

19%

5%

Knowledge Workers

74%

85%

Power Users

7%

10%

Source: Gartner (May 2013)

To calculate notebook TCO, we simulate two different usage patterns:

  • Day extenders — use the notebook in the office and bring it back home in the evening and weekends
  • Traveling workers (or "road warriors") — are on the road for a significant portion of their working time

They represent opposite ends of the spectrum in terms of usage patterns. Although many other profiles exist between these extremes, these two user profiles (see Note 1) help illustrate the TCO impact of different usage patterns.

The user mix adopted for travelling workers reflects a user population largely made up of executives, sales and other highly mobile professionals. Day extenders include a larger portion of knowledge workers and a smaller, but growing, percentage of task-oriented workers who have lower salaries and were previously served by desktops. The user mix affects a number of elements in the TCO. Perhaps the most obvious implication is on end-user costs, as the average salary impacts our assessment of the user's lost productivity.

Our calculations are based on an organization with 250 applications in all scenarios except locked, and that 45% of the applications are installed locally on the desktop, while 55% are server-based or browser-based, or otherwise clientless applications. In locked and well-managed PC scenarios, we figure the number of applications will be reduced by half, to 125 applications total. Reflecting the uptake of the application virtualization technologies used to package new applications, especially in the context of OS migrations, our model now accepts that 20% of applications are virtualized.

For managed scenarios, we assume that two servers (we used five servers in previous iterations of this research) will be required to host the management tool (one management server and one reporting server). Not all management tool servers can run on virtual servers; therefore, we note them as physical servers. To these costs, we add the cost of personnel dedicated to the management tools.

Figure 1 presents Notebook TCO numbers per user, per year. The direct costs are the sum of the hardware, software and facility costs, as well as the IT operations and administration costs, and represent the portion of TCO that appears in budgets. End-user costs, conversely, are hidden and don't appear in anyone's budget. However, we believe that measures taken to reduce costs include an analysis of the change to user productivity (see Note 2).

Figure 1. Notebook TCO 2013
Figure 1.Notebook TCO 2013

Source: Gartner (May 2013)

Overall, the TCO of notebooks shows a modest decline from 2011 of 2% to 3%, depending on the scenario (see Figure 2). Declining hardware and software costs have been offset by slightly higher salaries for IT and end users. Direct cost overall decline between 6% and 9%, depending on the scenario, is due partly to: (1) the lower complexity due to a higher percentage of hosted and virtualized applications, and (2) the adjustment we have made to our staffing ratios.

Figure 2. TCO Change From 2011
Figure 2.TCO Change From 2011

Source: Gartner (May 2013)

To measure the TCO difference between desktops and notebooks, we eliminate the end-user salary variable by creating a scenario with the same user mix for all platforms in the simulated deployments (see Figure 3).

Figure 3. PC TCO 2013 — Same User Mix
Figure 3.PC TCO 2013 — Same User Mix

Source: Gartner (May 2013)

Day extenders are, on average, 7% less expensive to support than traveling workers. Similar to what we see for desktop PCs, the biggest TCO savings are obtained by implementing user lockdown, as well as by using tools and having established policies and processes to manage systems. Day-extender notebooks managed in this way can achieve up to 43% savings, compared with an unmanaged day-extender notebook. Implementing user lockdown for traveling workers, however, is not recommended, and likely will not result in a TCO reduction of the same proportions. In this case, TCO will decrease 29% more than the unmanaged traveling-worker notebook TCO scenario. Total lockdown is typically not appropriate for users who work from a variety of locations and may need to change their settings or load drivers when support is unavailable. This is reflected in some cost items, such as Tier 1 and Tier 2 support, which will grow as a direct consequence of the lockdown as the number of calls to the help desk increases.

The overall TCO reduction that we see in a locked traveling-worker notebook environment can be ascribed to the impact of end-user operation costs: Since users are doing less for themselves, these costs go down, but users may not be as productive as they need to be. Generally, implementing lockdown on a population of notebook users is not trivial: The requirements of users should be well-understood and reviewed frequently.

Consider TCO and Productivity Implications of Replacing Desktops With Notebooks

Day-extender notebooks are currently 10% to 12% more expensive to support in a scenario where the user mix is the same for all profiles. Traveling-worker notebook TCO is 18% to 19% more expensive to support than desktops. In both cases, the implications are that organizations need to ensure they get enough additional productivity from users to make up for the higher costs of notebooks. In many cases, few additional minutes of productivity per week will be sufficient to recover the additional investment. This is why so many organizations are finding notebooks compelling purchases over traditional desktops. Desktops still make sense for users whose roles do not take them outside the office or for full-time teleworkers, and situations in which the use of notebooks cannot enhance the user productivity (see Figure 4).

Figure 4. TCO Comparison of Desktops and Notebooks at Various Levels of Manageability, Same User Mix
Figure 4.TCO Comparison of Desktops and Notebooks at Various Levels of Manageability, Same User Mix

Source Gartner (May 2013)

As the phenomenon of tablets is exploding, users are increasingly demanding IT to adopt and support such devices and are hoping that, in some instances, tablets can be their primary client computing device. (We advise clients to make sure that non-Windows devices can provide sound compatibility with their portfolio of applications.) Organizations should explore whether they can replace notebooks with tablets, and the TCO implications of doing so.

In some cases, tablets are all a user needs to accomplish his or her job, and tablets may add the convenience of instant-on and touch for users who need to operate the device standing or while moving. (See "Enterprise Tablets as Notebook Replacements: Limited but Growing" for a wider discussion on the trade-offs involved in replacing notebooks with tablets.) In some vertical industries, IT has successfully replaced notebooks with tablets that resulted in higher productivity, higher user satisfaction and lower TCO. Currently, however, tablets work as notebook replacements for a limited number of roles — for instance, C-level executives, sales and field services. The compromises required today make tablets unsuitable notebook replacements for most creative, nonroutine workers.

As in the majority of cases, tablets are still companion devices and their TCO will add to the TCO of the user notebook. This is why organizations struggle to build a business case to procure companion tablets for their users.

It can be argued that the overall usage pattern of a user who is equipped with a corporate notebook may be modified by the introduction of the tablet. For instance, if a traveling worker uses a tablet on the road, he or she may then use the notebook as essentially a day-extender device. Thus, the TCO equals the TCO of the tablet plus the notebook, but the notebook could move to a less-expensive TCO profile. In many cases, however, we still see workers equipped with notebooks and tablets while traveling, because they are concerned that applications they need while away from the office will not be available through the tablet.

For most organizations, tablets represent increased TCO, which is why many organizations are adopting or considering BYOD programs to accommodate users' requests. In a scenario where the user has a corporate notebook, but also uses a personal tablet through a BYOD program, the organization will bear the TCO of the notebook, as well as the cost of the infrastructure and labor required to accommodate, manage and secure the BYOD program. This ultimately will be less costly than equipping the user with two corporate devices, because the organization will not pay for hardware and hardware support. Furthermore, the organization often will not reimburse data and voice plans. We see this scenario as gaining increased popularity.

Figure 5 presents the TCO for tablets used and supported in two scenarios: platform and appliance (see "Use Managed Diversity to Support Endpoint Devices" and "Total Cost of Ownership of Mobile Devices: 2012 Update").

Figure 5. Tablet TCO
Figure 5.Tablet TCO

Source: Gartner (May 2013)

The platform scenarios represent a deployment where tablets are trusted, and procured and fully managed by IT organizations through each phase of their life cycle. Platform support for tablets is typically offered by organizations that deploy mobile business applications, which involves a more complex scenario from a management perspective.

In the appliance scenario, devices are untrusted and not necessarily owned by organizations. IT only supports a limited set of functionalities, typically (user) access to email, calendar, contact, browsing and potentially some Web-based applications. Appliance devices often use a series of isolation techniques that either keep secured content on back-end servers and deliver it through browsers or thin-client delivery models, or containerize the content on the devices within encrypted application bubbles on the device. The support requirements of tablets in this scenario are reduced and simpler, compared with other types of devices.

The TCO of tablets remains much lower than that for travelling-worker and day-extender notebooks (from 60% to 80%, compared with well-managed and locked-down notebook scenarios). The key caveat is that the tablet may be insufficient to serve the complete needs of the user and must be supplemented by other devices due to functional or compatibility needs:

  • For the notebook TCO, we account for a full end-user software allocation (inclusive of office productivity, collaboration, email client, browser, business applications and e-training modules).
  • For tablets, we consider that a basic suite of software will be bundled with the device.
  • In the platform scenario, we account for the IT labor required for managing one or more business applications, but we don't factor in the software and management cost of the mobile business application.

More complex is the question of when to use tablets and hybrid devices that also offer compatibility with Win32 applications (Intel-based Windows 8 tablets and hybrids). These tablets/hybrid devices combine the features of tablets with a full PC environment, which will be managed with traditional client management tools, with similar processes required to manage a notebook. Although we haven't published a TCO profile of these categories of products, we expect them to have a TCO that is close to that of a traditional notebook. Elements of uncertainty around the TCO of such devices still remain. For instance, we are still unable to determine the right set of configurations and pricing for a notebook replacement device, and the length of the life cycle (i.e., Will the life cycle of another device match or be shorter than that of a notebook?). These are all elements that heavily impact TCO. Organizations evaluating this category of tablets and hybrid devices as notebook replacements should look beyond TCO, because users may end up with bulky/heavy tablets or suboptimal notebooks.

Many organizations have two notebook standards: a large-screen notebook and a lightweight notebook. Tablets generally don't offer screens larger than 12 inches (diagonal dimension) and thus cannot serve the first standard, but they can serve the second standard with either a hybrid or convertible device, albeit at a higher TCO. Benefit calculations can mitigate this increased cost by avoiding the purchase of a companion tablet.

Be aware of the increased costs of non-Windows tablets that require the use of mobile device management (MDM) tools that are not as operationally efficient as the traditional PC management tools. MDM tools often require the user to be involved with the management process and have other limitations due to the architecture of the platform (see "Windows 8 Tablets Promise Manageability").

Note 1
User Mix: Traveling Workers and Day Extenders

Traveling workers often will use their notebooks in a variety of nonoffice environments, including hotels, airports and customer sites. Traveling workers were the first to be equipped with notebook PCs within organizations. They are often on their 3G or 4G notebook. Although they are the most expensive users from a support perspective, they are also increasingly astute users.

Day extenders demonstrate the simplest notebook computing environments to support. The simpler usage patterns of day extenders are reflected in lower TCO numbers.

Note 2
Chart of Account

For a description of each cost component, see "Client Computing Total Cost of Ownership, 2011: A Chart of Accounts." While each individual cost has been adjusted as described in this research, the general framework is still valid.