Self-service

When a loan officer is asked to "self-source"[10] they're taking on a responsibility that's not one of the top seven "Loan Officer Job Duties" listed by a major job placement service.[11][12]

A situation where no payment is made is self-service, such as airport check-in kiosks and checkout machines.[3] International borders have also experimented with traveler-assisted fingerprint verification[13]

Another situation is where a company's Human resources department is partially bypassed by departments that "source talent themselves."[14]

An early use of the term is a 2005 HRO Today article[15] titled "Insourcing, Outsourcing? How about Self-sourcing?" that mined Wikipedia's history of a pair of banks that merged decades ago as Standard Chartered and, after September 11, rebuilt its personnel department in an innovative way.

The concept is similar to self-service, and one USA example is pumping gas: New Jersey banned customers from doing this in 1949;[5] now NJ is the only state "where drivers are not allowed to pump their own gasoline."[5]

In 1994 it was considered a radical change to propose permitting self-service at the gas pumps, in Japan, and the New York Times reported that "the push .. (came) from Japanese big business ... trying to cut costs."[1]

Automatic Teller Machines are another example, and their expansion beyond banks have led to signs saying Access To Money, which refers to a company with that name;[16][17] the technology began over half-a-century ago.[18]

Selfsourcing is the internal development and support of IT systems by knowledge workers with minimal contribution from IT specialists, and has been described as essentially outsourcing development effort to the end user.[19] At times they use in-house Data warehouse systems, which often run on mainframes.[20]

Various terms have been used to describe end user self service, when someone who is not a professional programmer programs, codes, scripts, writes macros, and in other ways uses a computer in a user-directed data processing accomplishment, such as End user computing and End user development. In the 1990s, Windows versions of mainframe packages were already available.[21]

When desktop personal computers became nearly as widely distributed as having a work phone, in companies having a data processing department, the PC was often unlinked to the corporate mainframe, and data was keyed in from printouts. Software was for do-it-yourself/selfsourcing, including spreadsheets, programs written in DOS-BASIC or, somewhat later, dBASE. Use of spreadsheets, the most popular End-user development tool,[22][23] was estimated in 2005 to done by 13 million American employees.[22]

Some data became siloed[24] Once terminal emulation arrived, more data was available, and it was more current. Techniques such as Screen scraping and FTP reduced rekeying. Mainframe products such as FOCUS were ported to the PC, and Business Intelligence (BI) software became more widespread.

Companies large enough to have mainframes and use BI, having departments with analysts and other specialists, have people doing this work full-time. Selfsourcing, in such situations,[10] is taking people away from their main job (such as designing ads, creating surveys, planning advertising campaigns); pairs of people, one from an analysis group and another from a "user" group, is the way the company wants to operate. Selfsourcing is not viewed as an improvement.

Data warehouse was an earlier term in this space.[25]

It is crucial for the system's purposes and goals to be aligned with that of the organizational goals.[26] Developing a system that contradicts organizational goals will most likely lead to a reduction in sales and customer retention. As well, due to the large amount of time it may take for development, it is important allocate your time efficiently as time is valuable.

Knowledge workers must also determine what kind of external support they will require. In-house IT specialists can be a valuable commodity and are often included in the planning process.

It is important to document how the system works, to ensure that if the developing knowledge workers move on others can use it and even attempt to make needed updates.[27]

Knowledge workers are often strongly aware of their immediate needs, and can avoid formalizations and time needed for "project cost/benefit analysis"[28] and delays due to chargebacks.[29]

Additional benefits are:

• Improved requirement determination: Since they're telling themselves what they want, rather than someone else, this eliminates telling an IT specialist what they want. There is a greater chance for user short-term satisfaction.[22]
• Increased participation and sense of ownership: Pride and self-push will add desire for completion, sense of ownership and higher worker morale. Increased morale can be infectious and lead to great benefits in several other areas.
• Facilitates speed of systems development: Since step-by-step details preclude formal documentation, time and resources are concentrated, whereas working with an IT specialists analyzing would be counterproductive. Selfsourcing is usually faster for smaller projects that do not require the full process of development.[26][30]

Many knowledge workers involved in selfsourcing do not have experience or expertise with IT tools, resulting in: Pride of ownership has been found to be a major cause of overlooking errors. [31]^:p.30 A 1992 study showed that because Excel "tends to produce output even in the presence of errors" there is "user over-confidence in program correctness."

• Lost hours and potential: potentially good ideas are lost. These incomplete projects, after consuming many hours, often draw workers away from their primary duties.
• Lack of organizational focus:[26] These often form a privatized IT system, with poor integration to corporate systems. Data silos may violate policy and even privacy/HIPPA/HIPAA[32] laws. Uncontrolled and duplicate information can become stale, leading to more problems than benefits.
• Lack of design alternative analysis: Hardware and software opportunities are not analyzed sufficiently, and efficient alternatives may not be noticed and utilized. This can lead to inefficient and costly systems.
• Lack of security: End users, as a group, do not understand how to build secure applications.[33]
• Lack of documentation: Knowledge workers may not have supervisors who are aware that, as time goes on, changes will be needed and these compartmentalized systems will require the help of IT specialists. Knowledge workers will usually lack experience with planning for these changes and the ability to adapt their work for the future.[27]

Although departmental computing has decades of history,[21] one-person-show situations either suffer from inability to interact with a helpdesk[34] or fail to benefit from wheels already invented.[4]