Posted on Charity Village July 2002

Whenever we design an information system for an organization, we have to acknowledge that people are irrational, and that information never flows rationally through organizations. Technology initiatives often fail because their designers don't take into account how people actually learn, and who they are willing to learn from.

For example, in our study of settlement agencies in Ontario, many agencies told us that formal off-site training sessions were ineffective in learning new technologies. In contrast, small-group or one-to-one sessions that helped learners figure out how to use computers in their own offices, doing their day to day activities, were extremely effective. This is just one demonstration of a well known tenet of adult education; that learning is most effective when it is in the context of actual use.

But most workplace learning takes place outside formal training sessions. People learn through a variety of channels, including reading, going to conferences, trying out new skills on their own, and communicating with their colleagues. My previous article on knowledge management and social network analysis described how information flows through a network, and how social networks are critical in understanding how to manage knowledge.

Social networks are also key to understanding how to change and manage organizations. Two recent articles in the Harvard Business Review (June 2002) and the MIT Sloan Journal of Management (Spring 2002) summarize new research on the importance of informal networks in the way that work is accomplished. This research suggests that formal reporting relationships are only a small part of actual management, and that informal networks influence organizations through the relationships of people in four core roles. If managers acknowledge and understand who plays these roles, they can improve performance and also introduce change more effectively.

The whole area of social network analysis is growing rapidly, because it explains and describes so much of organizational life. In this article, I'll just describe the four key roles according to these two articles, as well as some of the myths of social networks.

The roles are central connectors (sometimes called hubs), boundary spanners, information brokers and peripheral specialists. Central connectors are people who seem to know everyone; they link individuals together in a network by knowing who can do what, and who knows what. They are the people you call most often for help, and if they can't help they know who can. They tend to spend large amounts of time helping colleagues with information (at least an hour a day), and if their role is not understood they may be undervalued in the organization.

Boundary spanners connect informal networks with external groups, for example other agencies, or sectors, or regions. Through their relationships with the outside, boundary spanners provide a quick efficient conduit of relevant information. They are especially important when agencies need to frequently tap into specialized skills, such as technical expertise or research knowledge. Sometimes long-term consultants can play this role in an organization. Organizations need to ensure that their boundary spanners are connecting to the right people inside the organization, so that relevant external information flows quickly to the network. For example, boundary spanners should connect to central connectors rather than people on the periphery of the network.

Information brokers connect several subnetworks within the organization, as opposed to boundary spanners who connect to outside networks. Information brokers are essential because they prevent the many groups within any large organization from falling apart into separate silos. They have many indirect connections throughout the organization.

Finally, peripheral specialists act as outsiders in the network, and are pulled in for specific advice or tasks. They are not tightly integrated into the group. They may not spend a lot of time communicating with others, and don't show up at staff meetings, but they may be investing their time in maintaining the necessary skills. For example, a visiting medical specialist may have a peripheral but very important involvement in a child care agency.

Organizations that want to introduce change will have a great deal of difficulty unless they work through informal networks and the webs of trust and communication that are already established. Tools like formal social network analysis can help to identify these relationships, and identify the real informal leaders in the organization.

**********

Gillian Kerr, Ph.D., C.Psych.

President, RealWorld Systems

gkerr at realworldsystems.net

Posted on Charity Village June 2002

Summary

1. Computers cost a minimum of $250/month per workstation including all direct costs such as networks, replacement, maintenance, technical support and user training. If agencies are not spending that much on managing their technology, they are probably spending more on the indirect costs of wasted time and lost productivity.
2. Funders should provide technology funding as part of program costs in the same way they fund telephones and rent.

**********

RealWorld Systems has just completed a study for Citizenship and Immigration Canada on technology needs of agencies serving immigrants and refugees in Ontario. Citizenship and Immigration Canada computerized the entire settlement sector in Ontario, spending about $12 million over three years to supply every funded agency with computers, networks, technical support and a state-of-the-art extranet. We were asked to make recommendations on how CIC should invest in the sector over the next few years.

We spent considerable time attempting to identify ways to support the technology requirements of these agencies as efficiently as possible. All of our analyses led us to the conclusion that agencies and funders should assume a minimum amount of investment per workstation in order to manage their technology effectively.

Further, we concluded that funders should provide technology funding directly to agencies as part of their operating expenses, based on the real costs of managing technology. It should be treated as part of overhead expenses, just like telephones and rent.

Estimating per-workstation funding needs

Research on “Total Cost of Ownership” (TCO)(1) has demonstrated that the initial price of a computer is a small portion of the real costs of supporting it throughout its life. Total Cost of Ownership (TOC) distinguishes between direct costs of computers, including hardware, software, repair, technical support and administration, and indirect costs, including downtime and loss of staff productivity. For more information on cost breakdowns, see the last section on 'Components of Total Cost of Ownership'.

RealWorld Systems recommended that per-workstation funding be based on direct costs only, since indirect costs are already covered under staff salaries. Indirect costs are 'hidden', because they are the hundreds of hours a year of wasted time as a result of computer crashes, email viruses and network problems.

Our best estimate of the direct costs of managing technology was a minimum of $250/month per workstation for office use.

According to Compugen, citing the Gartner TCO database(2), the lowest cost estimate for managing computer costs in an educational context is $2,500 (US) per user per year, including hardware/software, operation, and administration. This minimum cost reflects all possible best practices and does not cover end user operations or downtime.

A 1997 white paper by IDC analysing TCO in the educational sector estimated direct costs as $2,251 US. They stated that nonprofit organizations spend less on technology than businesses for several reasons; charity pricing of software, slower replacement of old computers, less powerful computers, and fewer technical support staff(3).

Our recommendation of $3,000 CDN per workstation per year is low according to published research, but is within the range currently being used by some innovative Canadian managed service providers, including Baudry Cybernomics. We believe that Canadian nonprofits need up-to-date research on best practices and TCO for their own sector but until then, $3,000/year is a good budget to work with.

It is important to note that the largest potential cost savings from the use of best practices are in the indirect cost category. In other words, a poorly-managed system may seem to cost about the same as a well managed system in terms of obvious direct costs, but will cause great amounts of wasted staff time and lost productivity.

It is also critical that agencies get competent Information Technology (IT) support. Those who can't afford to hire their own IT staff should hire external consultants or part-timers (using their budget for direct IT costs) and assign an internal manager to oversee the technology services. Technology management is not easy, and most agencies will not have the capacity to do it alone.

Components of Total Cost of Ownership

Compaq Corporation's Total Cost of Ownership Snapshot Report(4) defines the TCO approach and the cost components that are generally included. The following material is directly excerpted from Compaq's report, with the exception of comments by RealWorld Systems in italics.

The Total Cost of Ownership (TCO) of a computer system is defined as all of the costs associated with the asset over its entire lifecycle, i.e. acquiring, using and supporting all information technology (IT) assets. TCO includes the cost of all people, processes, and technologies associated with the use of that asset. TCO specifically refers to the cost of owning and operating networked information technology assets: desktops, LANs, and servers in a client/server environment.

TCO is split into two types of expenses: direct and indirect.

Direct IT expenses are those items that are typically budgeted to the IS cost center, such as hardware, software, management labor, operations labor, development and communication fees.

• Hardware and Software (capital expenditures and lease fees for new installations, upgrades, and updates).
• Operations (network, system, and storage administration labor and outsourcing fees, reactive and proactive management tasks).
• Administration (helpdesk, training, purchasing, travel, maintenance and support contracts, and overhead labor).

Indirect IT expenses are those items that are not budgeted and often go unaccounted for in most organizations including end user self and peer support, casual learning, and productivity losses due to downtime.

• End user costs (peer and self support, casual learning, and “futz”).
• Downtime (lost productivity due to planned and unplanned outages).

Cost Categories included in Total Cost of Ownership:

Budgeted (Direct expenses) These costs should be included in per-workstation funding
HW/SW Costs: Hardware: Expensed, Depreciated or Leased Upgrades Spares Supplies Software: Personal Productivity and personal database applications Business and engineering software Database, management and development tools Messaging and groupware Other software Network, systems, storage and asset management software Service desk management software Training software	Operations Costs: Technical Services – Clients, Servers, Network: Tier II problem resolution Tier III problem resolution Traffic management & planning Performance tuning User administration (local & logical A/M/C) OS Support Maintenance labor Software deployment Application management HW configuration / re-configuration HW deployment Disk and file management Storage capacity planning Backup and archiving Repository management Planning and Process Management Account management Systems research, planning and product management Evaluation for purchase Security and virus protection Business recovery Database Management and Administration Service Desk (Tier 0/1)	Administration Costs: Finance and Administration Supervisory management IS administrative assistance Asset management Budgeting and chargeback Auditing Purchasing, procurement & contract management Vendor management IS Training IS course development IS training delivery End User Training End user course development

 

These costs are covered by regular staff salaries and are often 'hidden'. They represent the greatest potential for savings.
End User Operations Costs: Peer support Casual learning / support Formal learning File and data management Application development Futz factor End user metrics Time seeking peer support Activity while waiting for resolution	Downtime Costs: Productivity lost to outages of resources

Footnotes:

1. See especially the following web sites for information and best practices on TCO analysis: techguide.merit.edu/plantemoran1.htm, www.classroomtco.cosn.org, www.compaq.com/tco/bestpractices, and www.educause.edu/ir/library/html/cem9829.html. All of these resources are US based. Compaq offers a software tool to estimate the extent to which an organization is using best practices in managing technology costs and performance at http://h1800.www1.hp.com/tco/snapshottool.html – it's highly recommended. Gartner is the leading research and consulting firm associated with TCO analysis; see www3.gartner.com/4_decision_tools/measurement/decision_tools/tco/tco.html.

2. Personal communication, Maarten Verhaar, Director of TCO Practice at Compugen: See also www.compugen.com/services/tco.htm#.

3. “Understanding the Total Cost and Value of Integrating Technology in School”, a white paper sponsored by Apple Computer, 1997, International Data Corporation, www.hubster.com/apple/whymac/idc-tco-white-paper.pdf.
4. Compaq Computer Corporation. Total Cost of Ownership Snapshot Report (1999). ftp://ftp.compaq.com/pub/solutions/tco/TCOToolSampleReport.doc. (Please note: clicking on this link will download a Word document to your computer.)

************

Gillian Kerr, Ph.D., C.Psych.

President, RealWorld Systems

gkerr at realworldsystems.net