Failed projects mean squandered resources and investment, even if
parts of the work undertaken can be re-used. Karl Cushing hears how
a panel of old hands would dam the spate.
Can you say in all honesty that every IT project you have steered
has been an unqualified success? Most major implementations fail in
one way or another and result in wastage of resources. IT project
failures are costing the UK millions of pounds, according to
analyst organisation Giga Group, and at least two-thirds is
avoidable.
Because of the time and scale involved they are often out-dated by
the time they are completed and fall foul of management churn.
Companies should maximise the chances of success by planning more
carefully, breaking down large projects into bite-size chunks and
ensuring that there are lots of achievable short-term project
milestones and early deliverables.
This at least was the consensus view of those attending a recent
roundtable event, held at Brunel University, Middlesex, on finding
ways to stem the "flood" of money and resources wasted on
unsuccessful IT projects.
Much of the time the battle is lost before it has even begun, with
companies failing to ask themselves basic questions such as what do
we want to achieve and why are we doing this. According to Tony
Morgan, independent consultant and visiting research fellow at
Brunel University, there should be more of an onus on the business
side to identify what it wants to achieve through the project
before the IT department gets on board.
Failure to define strategic objectives, combined with poor
communication, results in wasted time and effort through
programmers coding the wrong things, says Morgan. He argues that if
they were properly briefed you could eliminate 80% of the problems
on the IT side. Automating more of the process and using a
model-driven architecture can help avoid IT failures, he adds.
According to Simon Ratcliffe, a senior IT consultant at Business
Systems Group, too many projects are borne from a misplaced desire
to keep up with rival companies. "Pointless re-invention" results
in projects going too far and getting rid of good process design as
well as what is bad. "Companies need to be agile but not radical,"
he says. "After all, people aren't radical."
Mundane projects can deliver easy savings wins. Aiden Walsh,
vice-president for IT at pharmaceutical firm Quintiles Europe,
points to the huge value that projects involving the
de-commissioning of legacy systems can provide.
Another common mistake is to underplay the human factor in a
technological implementation. Paul Bevan, director of strategic
marketing at Unisys, says companies often end up de-skilling their
staff by installing "intelligent" technology when what they really
need is intelligent staff using intelligent technology.
Mark Lycett, project leader of the fluid business research project
at Brunel University, agrees, adding, "All too often business
systems see people as technology slaves or process followers and
not as problem solvers."
The need for IT and business to work closer together is a constant
theme. Morgan stresses the need to create a common language so that
the business and IT functions can understand each other. The
present lack of communication creates a lot of scope for passing
the buck, he says.
The IT department should also be more proactive and willing to
assume the leadership role in IT projects instead of simply
answering questions from the business side. "We need to have the
courage to say that this is what we can do for the business," says
Ratcliffe. "A lot of boards haven't got a clue."
Buying packaged software instead of developing your own can also
help to reduce waste. Whilst Morgan says many companies simply opt
for the off-the-shelf option because it is easy, Ratcliffe argues
that most companies are more like their competitors than they think
and so buying packaged software often makes sense. Walsh agrees,
saying that this desire to do things yourself often ends up in a
confusing, spaghetti-like mass of systems and interfaces. "We
sometimes over-complicate things," admits Peter Crutchfield, head
of IT at Marie Curie Cancer Research.
Projects go wrong for a whole myriad of reasons, but Lycett says
managers should not be afraid to pull the plug on a project: it can
show that you're moving with the times and can avoid further
wastage. The important thing is to make sure that the reason for
the failure isn't down to you.
The fact is that a lot of IT waste is avoidable and the potential
for savings is enormous. With IT budgets under the spotlight and
margins ever tighter, this is an area that cannot and must not be
neglected.
Why do large projects fail?
- Too much needless re-inventing: if it isn't broke, don't fix
it
- Projects initiated for the wrong reasons
- Business and IT not communicating
- Too many projects aimed solely at integration and don't really
add value
- IT staff are afraid to ask questions.
How can you avoid waste in IT projects?
- Plan carefully: why are you doing this and what do you want to
achieve?
- Break large projects down into smaller bite-size chunks
- Have internal milestones and early deliverables
- Don't be afraid to cancel projects
- Senior IT staff should take more of a leadership role in
technology implementation projects
- Don't develop just for the sake of it: packaged software might
do the job just as well
- Don't underestimate the importance of staff. /li>
Who was at the round table
Dr Mark Lycett, head of the fluid business research
project at Brunel University
Spent 12 years working on
feasibility and development projects before getting involved in
research. A fluid business is defined as agile and integrates
people, process and technology to enable it to respond to its
changing environment with limited disruptions to its operations.
Tony Morgan, independent consultant and visiting research
fellow at Brunel University
Author of Business Rules and
Information Systems: aligning IT with business goals. Prior to his
current role, Morgan was the senior solutions specialist at Unisys
and headed up information system development projects in the UK,
Europe and the US.
Simon Ratcliffe, senior IT consultant, Business Systems
Group
Responsible for the planning and design of large
infrastructure projects, including networks, systems design,
messaging and workflow and specialises in designing and installing
financial reporting systems.
Peter Crutchfield, head of IT, Marie Curie Cancer
Research
Worked in service delivery, infrastructure and
development and has been a lecturer in information systems. Prior
to taking up his current role, he was responsible for a number of
substantial IT projects in the City.
Aiden Walsh, vice-president of IT for contract
pharmaceutical company Quintiles Europe
Responsible for
the IT strategy and operations of Quintiles Europe. Previously he
was vice-president for corporate IT at GlaxoSmithKline.
Paul Bevan, director of strategic marketing,
Unisys
Held a variety of sales and marketing roles with
Unisys since he joined the company in 1984.