Death of the IT project

Dharmesh Mystri, CTO, edge IPk, explains why business driven integration will resuscitate failing IT Projects

Change is inevitable in every business and the rate of these changes is increasing mainly through innovative technology.

However, if technology is driving these changes, why is it that the way we deliver software applications has largely stayed the same for the last 70 years? The same structured approach to software development has been used since the 1930's when Bell Labs' Walter Shewhart proposed an approach called "Plan, Do, Study, Act", PDSA.

In recent years the cycle for software development for most companies has been described as the Waterfall.

This is a sequence of activities that starts with a business defining its requirements for IT automation (Define), communicating these needs to an IT developer (Design).

The System is then built (Develop) tested (Test) and finally rolled out (Deploy).

Each of these steps follows on from the last, hence the term "Waterfall".

Despite huge investments in training, tools and methodologies, failure rates of IT projects remain extremely high.

Statistics vary between analysts but typically suggest that 66% of projects fail and 30% of projects get cancelled even before the development stage.

In the 90's more agile methodologies started to appear, starting with Rapid Application Development (RAD) - an approach that promoted prototyping applications.

Whilst these accelerated the development of systems generally these prototypes had to be discarded.

Even when these prototypes were "evolved" to actual systems the underlying architecture of the solution in over 85% of these approaches meant the changes for new requirements were hard to implement.

Therefore systems built quickly tended to be more expensive to maintain and change.

Whilst there are many issues relating to these failings of these IT projects the underlying issue is that documents do not describe systems well, and the time between specification and seeing a solution is too long.

The answer surely is about getting business users to create applications for themselves, but due to the complexity of integration this is a tough goal to achieve with today's technology.

However, in terms of specifying a system, generally a business user is focused on what other users will see and use rather than the integration.

If end-users can use a tool to enabled them to create applications and GUI's without having to write programming code, then applications could be developed more accurately for the business need and far faster.

And if the same tool could be used by IT for integration and ensured the end solution was created in a way that future changes and maintenance was easy they would have an approach that will provide agility, efficacy and flexibility.

The three simple steps to a successful business driven project are:.

1 Business users model application screens.

2 IT provide integration.

3 The tools ensures a flexible architecture for change.

This approach not only produces the right result but also gets IT and business working closer together.

Getting this right couldn't have come at a better time because according to Butler Group "Today businesses conduct mission critical activity through the Web".

"This trend has also increased the amount of application development on Web-enabled front-ends".

"For large enterprises the effort is enormous, the work to integrate applications adds further strain on development teams".

"The answer has to be better use of automation and tools that reduce development time of building applications.".

Back to top