Wilder world predicted for IT project management

Bringing in projects on time, under budget, that satisfy the customer is becoming more difficult than ever.

A group of cheetahs together is called a coalition, according to Gartner at its latest summit on IT governance, and project and portfolio management. Make of that what you will. However, one might almost imagine the research giant was suggesting that mass collaboration, by its nature, might encourage a lot of individuals to make more speed by cutting corners. Haste makes waste, as the old saying goes.

According to Gartner, IT projects are handled today in ways that end up not meeting the needs of the customer organisation. Worse than that, the methods used are tending to result in even more costs being forced on to the shoulders of the long-suffering business customer.

Andy Kyte (pictured, right), vice president and Gartner fellow, gave a keynote at the event explaining that judging IT project success requires a long-term perspective that is sadly lacking in many project teams. "I'm a bit like Daniel in the lion's den on this," he said. "I have a message from the research we do with clients -- from all aspects of technology in business -- that may be slightly unpalatable."

IT teams have sets of criteria they use for judging the success of their projects but these criteria may not match the goals of the customer. The result, far too often, is that the IT team declares the project a success although the end user does not, he said.

"What happens after the project has finished?" said Kyte. "The success of IT projects should not be measured by how successful the IT applications are [as applications]."

Change management
For example, if an application needs to be owned and managed for 15 years, how easy is it to make changes to that application? If the end result of a project is an application that needs to last 15 years but is costly to update or make changes, the total cost of ownership (TCO) becomes much higher. Yet that consideration is rarely factored into project parameters from the beginning, suggested Kyte.

"Go-live costs are only a percentage of TCO over the 15 years. And we looked at the data through many different filters and internally came to a realisation," Kyte said. "How much will the app change over its life? If it doesn't change very much, more of the TCO will be involved in the initial project. And if it's highly volatile, you are likely to spend a lot more."

One factor, he said, is the design objectives. Often there is far too much focus on features and functionality, and very little - if any - attention paid to the more important requirements as far as the customer is concerned, things like availability, flexibility, usability, stability, and maintainability.

"For example, designing for small and simple programmes - rather than massive compile units with many lines of Java, et cetera," Kyte said.

Customers are likely to want an app to have low TCO, to be easy to support, easy to maintain, and easy to extend, he said, and the engagement between the IT team or project management and the customer needs also to be managed in a way that reveals what the customer's needs truly are. That also often doesn't happen, he said.

"Nike's slogan, ‘just do it', is in fact the most used project methodology," Kyte quipped. "Or I want it fast and I want it cheap."

Many software projects are complex and on-going. CRM projects, for example, are notoriously long-term and often never ‘finished' per se, he noted. So although defining the scope of the project remains critical, an effective and useful project definition may be far less about the granular detail of application functions and features than has often been supposed.

Also, something that looks successful perhaps at the customer organisation's board director level may look quite different from the perspective of the HR department, or that of the end user actually using the new technology day by day. For that reason, project teams should expend much more effort on defining the project requirements and part of that is about getting more different inputs from a broader range of stakeholders on what is needed, he suggested.

"If you spend just six per cent of the TCO on the actual project, the customer will spend 94 per cent during the rest of the life of the system," said Kyte. "The problem is that you [the project management team] don't have a clear understanding of what is meant by ‘requirements'."

He said that IT providers should also refrain from "selling it thin to get it in". In other words, project managers are not only glossing over details that might affect the TCO but calculating a lower budget that would be more likely to tempt purchasers -- although the cost would probably need to be revised upwards as the project progressed.

Too often, he said, the original pricing that was presented to the customer during the bidding process is revised four or five times - and always upwards. This cannot help but leave a bad taste in customers' mouths and make them wary of signing up for IT investments, Kyte suggested.

Today, CEOs and CIOs don't talk about reliability as the holy grail of their IT project but about the time it takes to implement changes. Project teams would do well to remember that, he hinted, and at the very least should expend more effort and analysis on the ‘requirements capture' part of the process.

Matt Hotle (pictured, left), vice president and Distinguished Analyst for application governance research at Gartner, agreed, saying that PPM as much of the IT industry currently understands it must end. Project management must become much more agile and holistic in its approach, and less about delivering a set piece of software in a set time.

Business process management must be a key focus, to improve efficiency and avoid layering ever-more onerous processes on top of others when those processes add nothing to the customer's core business and simply reduce productivity.

"Business people are saying that we [IT PPM] are too slow, too expensive, unreliable and not meeting their business objectives," declared Hotle. "And then they say, ‘well, I'm going to go and build my own SaaS product instead'. And then we will have to live with that."

Applications are much more expensive to maintain than to build, confirmed Hotle. Because of this, IT providers should take a leaf out of the big-brand vendors' product roadmap book - looking at the whole lifecycle, changing customer needs, and future iterations of the IT project instead of simply developing something that meets an immediate need.

This transformation is critical for IT providers and it is already starting to happen, so project teams should already be considering their next moves, Hotle insisted.

He also said teams should be more siloed and rely more on permanent staff with a genuine stake in the IT project or business itself. Currently, a succession of temps tends to do a lot of the grunt work on development and support as well as manage projects. Often they work on similar tasks across projects - for example, a messaging specialist may be carrying out the messaging tasks on a series of projects.

If you won't have to fix a messy piece of code yourself or work with it in any way, you will be less motivated to create an easy-to-support product in the first place, Hotle pointed out.

Continuous evolution
Gartner believes that the very pervasiveness of IT is intensifying business change, forcing businesses to adapt in ways that IT organisations are not currently structured to support. This dynamic is being felt most where IT-intensive functions and services are tied tightly to business success. Different siloes - such as PPM, enterprise architecture, application delivery, IT services management and operations - need to act in concert, fast enough, to address these changes.

By 2016, IT organisations will be increasingly scrutinised and expected to deliver in an integrated way, in near-real-time, to help business customers. Demand for agile project leaders with project and change management skills as well as domain expertise will increase 75 per cent, according to Gartner, and jobs for pure-play project managers will become increasingly scarce. And many IT organisations will reach breaking point when it comes to resources by 2015; there may be no future for those who cheat.

More information is in Gartner's report "Predicts 2012: PPM Leaders Must Embrace 'The Next Normal' by Embracing Changes and Adapting".