One of the world’s largest-ever construction projects, it was built out of what was formerly ocean. The airport opened and - lo and behold! The computers foul up and $3 billion worth of freight is turned away!
Does the construction industry know something about project management that the IT industry doesn’t? Is there some black magic about project management that has not been passed on to software engineers?
What secrets could there be? History’s greatest project manager must be Hannibal, who led an elephant-mounted army of disparate nationalities across North Africa, through Spain and over the Alps before giving the Roman Empire’s finest a right pasting.
The Egyptians could organise a few people in co-ordinated fashion too, as several large geometric structures in Giza testify. What is the average software project compared to that?
Surveys routinely indicate that some 70% of IT projects are either late, over budget or don’t work.
Of course it is the in-house IT people who are to blame. It couldn’t possibly be the suppliers because all their hardware and tools are so easy to use. Far more comfortable to assume is that failure is the fault of software project managers, all of whom should be sent to bed with no supper.
Project management is easy. After all there are countless excellent books on the subject. The Right Way has been set out. Can they not read?
Many Projects fail because it’s too expensive to do it properly.
For example it's is well-known how to write software so that it doesn’t fail. It’s just very expensive to do it that way. People don’t pay for what it costs: they pay for what it’s worth. Software that is part of complex fly-by-wire planes does not often fail
IT in some ways is an unusual industry because people deliberately set out to write software that they know is not going to work because the cost of being 100% sure that it will work is too high.
People have a remarkable tolerance for computer failures: you read about them all the time. You go to check in at an airport and the computer is ‘down’. How often does your Windows system break down and yet you are still willing to pay Microsoft money for it?
Software is called software because it is soft. Because it is soft you can change it easily and because you can change it easily you can get yourself in a mess. What makes it most attractive to mankind is the very thing that makes it difficult to manage.
It may be a comforting thought that software will never work perfectly. After all, how can a machine take over the world if it crashes and needs rebooting twice a day?
But there may be lessons to be learned from other industries which might at least improve the process.
It is arguable that the IT industry is about 30 years behind the construction industry in project management.
There have been considerable improvements in the 10 years and people have seen the error of their ways! New method, certifications and qualifications have helped, PRINCE2, Agile, PMP have all contributed to a wider acceptance of project management as a discipline.
Projects used to be run by techies but there is a growing acceptance that businesses need professional project managers.
One of the biggest problems in the industry is the way that contracts are written. In the construction industry in the old days the contracts were written so that if you didn’t perform on time then there were penalties.
These penalties were at the very least the profit and sometimes the value of the contract. Accordingly, the construction company made sure that whatever happened it wasn’t their fault that a situation had been taken down the route of conflict from day one.
These days building tends to be released before the scheduled date rather than after - and part of that is the adoption of a partnership approach. They tend to get a bonus for completing early rather than a penalty for being late. If the project is going wrong toward the end, and you are going to have to pay a penalty, then there is no incentive to put more money into it. If there is a bonus then there is an incentive to go for it.
The way the contract is written sets the scene for how the project will run. If a customer decides he wants a fixed price then he thinks he knows how much it is going to cost him. But things change and technology changes.
If you originally said you wanted it red and now you want it yellow they will say ”you can have it yellow but it is going to be expensive”.
There are companies out there who will accept a contract on any terms and then make sure that they get their money. But it’s not necessary to take this approach and you can structure the contract so that both sides can win.
Another problem is that the industry takes people with technical skills and assumes they can run a project. Do you think the project manager who built the Channel Tunnel knew everything about the technical details? You have people in your team who specialise. In our industry you would have people who specialise in the technology, but what you need is someone running the project who can run a business, the project is just a business that runs for a limited time.
Can you imagine an aircraft company where its entire future depends on a $10 million development project leaving it totally to someone with no project management training?
Project management is not about using Microsoft Project and coming up with the plan – it’s about dealing with people, handling conflict, negotiating with your boss and with your team. It’s about leadership as much as management and most programmers are not capable of doing those things.
When you are managing a project you will spend most of your time moving desks and PCs around, or else you are trying to stop people stealing members of your team to work on maintenance.
I’ve often ended up in situations where I have been fighting to get adequate resources and stopping other people poaching my team, so that people have said to me “you’re prepared to do anything to make sure your project is a success, aren’t you?” As if somehow this was a bad thing.
Many books set out the correct procedures for software project management, but they assume a perfect world.
Technology is often the least important factor in the success of a project. Nor is it about the project management tools.
The problem often starts when the project is first defined: Clear definition of KPI's, scope, business benefits all play a crucial role in ensuring that the project starts on firm footing and hence stands a good chance of success.. and of course a clar statement of the benefits the projects expects to deliver since at the end of the day benefits are the whole rationale of the project...
3 comments: