Quantify and control your 'underground'

Staff will always adapt software programs to suit their needs, but IT must be informed and consulted

Staff will always adapt software programs to suit their needs, but IT must be informed and consulted

It is no longer surprising that much more IT development takes place outside corporate IT departments than within. The growing PC population, disenchantment with delivery priorities, and increasing numbers of "power users" have created an enormous and mission-threatening challenge to many IT directors and organisations.

It happens for the wrong reasons: disenchanted with out-of-date online applications, and not understanding the IT project submission process, users develop local solutions to meet their needs. They are proud of them and share ideas with colleagues until it quickly becomes a strategic application that is enhanced all the time.

It also happens for the right reasons: when IT departments are deemed to have little understanding of the business; when there are technology projects; or when companies refuse to be driven by technology.

"So what?" you may ask. "We supply the machines, the easy-to-use applications and, if business people choose to develop their own solutions, that is relieving the burden on IT departments. Also, PCs are integral to many people's work." But there are compelling arguments against allowing ad hoc developments.

People who develop local solutions rarely have any documentation, so the original developer quickly becomes the help desk, which is fine until they leave the department or the company, or it becomes too much for them. Such developments seldom follow standards or procedures, and ownership is unclear - until something goes wrong.

Business people should be persuaded not to spend time playing with applications when they should be doing the job they are employed to do. Putting such applications on local area networks can bring down other systems.

Meanwhile, word will quickly spread that IT is slow at doing things compared with Mike in marketing, who will run you off an Excel macro in two days. The hidden costs associated with this trend are enormous. These applications must be explained, supported and even integrated with existing systems.

Company structure has an impact on the ownership, control and priorities of IT projects. The one-company approach, with one IT department serving all, has as many advantages as disadvantages. Strategic business units each with their own IT services and support have the opposite.

The key here is not to pretend one way is better; it is not. The challenges and consequences must be understood. Failure to consider them could consign the organisation to follow one approach for a while until business managers complain about poor focus or local delivery and switch, or operate in "silos" for a few years until the company realises there is too much duplication in costs and resources, and too many conflicting standards and approaches.

There is no easy solution. However, you can minimise problems by:

  • Standardising on one office suite
  • Setting up a group of senior IT/business users to document the number of user-developed applications
  • Auditing each application, record who owns it, supports it, and check documentation
  • Establishing how many people use the application - is it mission-critical?
  • Prioritising your mission-critical exposure
  • Having the IT department put standards in place for such development
  • Estimating the total cost of owning and supporting these applications, and share this information with business users at all levels.

This subject goes to the heart of the "who drives the future?" debate. Technology is there to be used but it must be acknowledged and quality controlled.

To join the debate e-mail [email protected]

Read more on Business applications