While 59 per cent of mission critical applications are developed from scratch, in 1996, US IT shops wasted $82 billion on IT projects that were never implemented, according to the Standish Group consultancy.
In the same year, only 27 per cent of all US projects succeeded, 33 per cent were ?challenged? and the bulk, 40 per cent, failed. A lot of client/server initiatives were also cancelled due to uncertainty about the potential of the Internet, but the average project cost fell from $2.3 million in 1994 to $1.2 million due to a 65 per cent reduction in their scope.
Speaking at the Fifth ABT Project Leadership Conference in San Francisco this week, Jim Johnson, president of the Standish Group, said the most common reasons for project success were user involvement - which was unheard-of five years ago - executive support and the role of the project manager, who is considered a key component in the mix.
He explained: ?There?s a better than 50 per cent chance of success if you keep the project below $750,000, take no longer than six months and have no more than six people on the job. If it gets too big, project managers can?t communicate or do anything else - they have to maintain balance."
Simplicity is also a help. "Keeping the project simple, with a minimum of people and open communication lines works best," Johnson went on. "It?s also a big plus to build your application on a standard infrastructure. Mix this with a good project manager, tools to keep track of the project, iterative processes and staff adhering to key roles and the project will work.?
Barbara Cooper, group vice president of information systems at Toyota Motor Sales, believes that the only way to succeed is to ?channel the chaos? and manage complexity in the best way possible by aligning IT with business goals and a clearly defined business strategy.
As a result, one of the main roles of the project leader is to understand and believe in the project goals and translate them to the IT teams, business executives and users.
Johnson explained: ?Project managers are the driving force between success and failure, but they have to have clear business objectives, know where the project is going and what it?ll look like in the end. They should control the flow of information between users, executives and technology and be the focal point. They need business and technical know-how, judgement, communication and writing skills, organisational skills, time management and diplomacy. In fact, the softer skills are more important than hard technical skills.?
According to Ed Yourdon, chairman of Cutter Consortium, project managers should also be good planners because the negotiations undertaken at the start of the project will determine whether it works or not.
?Afterwards, you?re just rearranging deckchairs on the Titanic. But, make sure negotiations are interactive. If the customers pushes in one parameter, you?ve got to show another will be pushed out,? he said.
Rhonda Hyacinthe, manager of the project office at Motorola Computer Group, agrees that planning is the foundation to managing the project and making sure it is delivered on schedule.
?You need a starting place and know how to start the project," she said. "The boss often has a good idea, but no formal way to start, so you need to establish the budget, the scope, resources or staff and then announce the project, so everyone involved knows about it. If you don?t know what the customer wants, you can?t deliver what they need. Also, if the planning is not good, it will be difficult to put in motion what you?ve planned for."
While it may be impossible to provide all the requirements in one project, it is imperative to document the aims and how they are to be undertaken to be able to identify and manage risk, she added. It is also necessary to create a schedule, a delicate task in itself. If it is too detailed, people will feel hounded - if not detailed enough, the project manager will not know what is going wrong until it is too late.
Andrew Gay, chief executive of Drake and Scull, a UK provider of electrical services, warned, however, that objectives do tend to change during the course of a project.
?Always remind the customer that 'scope creep' means profit and time seeping out the door. Change causes problems with technical and financial issues, timing and resources. A project manager must be in charge of every decision, but always ask why, what is this doing for the customer. Some 98 per cent of projects go wrong because people try to give that little extra and lose focus,? he explained.
Standish?s Johnson also pointed out that it is best to start with small iterative processes and move forward because growing software projects seems to be the key to project success. Project managers have to learn the word 'no', but should ensure they produce deliverables periodically.
This keeps the IT team focused on near term goals, rather than getting lost in the mists of a distant grand scheme. It also keeps business executives happy because they see they are getting something for their money.
But, according to the Cutter Consortium?s Yourdon, one of the most important things to learn is the ability to say no, combined with the ability to know when a project is dead and when to kill it in order to save time, money and resources that could be deployed elsewhere.
He explained: ?If the negotiations fail, just say no - you can?t do the impossible. If you can?t say no, appeal to a higher authority, for example send an email to the CEO. With the Year 2000, if the result of failure is bankruptcy and if the failure of the enterprise is at stake, an email is justified. You may make enemies and burn bridges, but the industry is short of 300,000 IT staff, so it?s not the end of the world. Don?t be pushed into an arbitrary deadline and negotiate irrational trade-offs or the project is doomed from the start.?
However, project managers need to be aware that their projects will be subject to change as they go along and a risk management strategy should be implemented to deal with this, even before the project begins.
Drake & Scull?s Gay recommends that no project should have more than five or less than three key items of risk and value.
?Identify risk early, discuss it with the customer and frequently review the project. Don?t let the fluff of lesser issues pull the project down. If a jugular issue shows a rash assume the disease is potentially fatal and panic early. Call a crisis meeting because you can still do something if you do something early. If in doubt, call time out because it?s your career on the block,? he said.
Tim Lister, principal of software development consultancy, the Atlantic Systems Guild, believes that to undertake successful risk management, managers need to be problem anticipators rather than problem solvers and should introduce 'tripwires' so they can recognise problems and diagnose them early on.
?Potential problems are inherent in the business, so accept risks because they?re tied to benefit. You don?t hold all the cards, so you have no control over some things. Any risk is a potential problem and the best laid plans may turn sour. First identify the risks. Once you?ve identified direct risks such as cost, you can take follow your action plan, but with indirect risks such as performance that are out there looming, you need a mitigation strategy to say if this happens, I will do that,? he explained.
As a result, another key skill in the view of Toyota?s Cooper is good communication skills.
?You need to know how to use multiple language forms to explain the same thing. We all speak English, but have different interpretations of what?s been said and expected, especially with business sponsors versus IT teams. Start with a lexicon and define terms, get everyone in a room and spell it out as an issue and get people to give feedback,? she explained.
But, she continued: ?Don?t use overkill - customise your message and know your audience. Use repetition to reinforce the same messages and communicate immediately what the future might look like. Provide a set timeframe and deliverables and ensure people understand what is to be delivered. In fact, have a dedicated communication analyst on each project to ensure communication flows smoothly between the IT team and the rest of the business,? she added.
CSC?s Bill Domeika believes that aligning IT with the business strategy is key because a project manager needs agreement between IT and business on what the goals are. The two should set priorities together and ?both have skin in the game? to ensure everyone takes responsibility for the project.
?Get IT to participate in the strategy process any time the business processes are being defined. Develop the business case for change concisely and succinctly. Change has an impact, but it?s necessary that people understand the benefits. And measure any project for the strategic and economic value,? he says.
Each project also needs an executive sponsor, who should be the driver and visionary and know where the project should go. It is his role to eradicate the political landmines that all too often bedevil projects and take responsibility for it. Users, on the other hand, need to understand the business and its needs and be there when necessary, balancing the fear of change with change for change?s sake.
To facilitate this exchange, Toyota?s Cooper recommends that project managers introduce a project office or institutionalised logistics unit to manage and oversee things across all ongoing projects. This should manage skills and systems resources to regulate the flow of supply and demand and coordinate peers, staff and supporting executives.
Project managers should also have the right to choose their own staff or at least have the right of veto, although, according to Yourdon, they are all too often ?just sent rejects and malcontents?. The problem is then compounded because few companies are imaginative enough in their recruitment processes.
But, the team is important because ?you can?t take a ragtag bunch of mediocre people and turn them into a star team. Don?t just focus on building a technical team though, look at team roles. Motivation is also important, so find ways of motivating people. Money is not always possible, so find substitutes such as training or using advanced technology, but act as a buffer to cut out the administrative nonsense,? he added.
This includes determining which administrative rules are worth breaking to boost productivity. Actions such as turning off phones and not responding to interruptions can increase productivity by 260 per cent, he claims, while Jim Loehr, president and CEO of LGE Sport Science believes that healthy and happy people tend to produce better.
In conclusion, Yourdon also attested that well chosen tools, including project management tools, prototyping, rapid application development, configuration and version control products, can help tremendously, but are not used in 90 per cent of projects.
Instapaper to 'go dark' in Europe until it can work out GDPR compliance
James Robbins of ArrowXL says that AI is no longer 'tomorrow's technology'
Staff told to beware of "unusual sounds" after an employee reported mystery symptoms
Sophisticated malware comprises code previously used to attack Ukraine