The Cost of Failed IT Projects

Failed IT projects

Failed IT Projects are costing companies and countries an absolute fortune every year. The cost of Failed IT Projects accumulated each year comes to more than the Gross National Product of some hefty sized countries.

Let me first of all give you some shocking facts related to the development of technology projects.

EU figures looking at 214 technology projects from 1998 to 2005 showed that only 1-in-8 technology projects met their time, budget and quality objectives.

Indeed the total cost of technology project overruns in the EU in 2004 was €142bn – yes, that’s billion.

They found that 23.8% of all projects don’t get finished at all.

Project Failure

A survey by KPMG in 2010 showed that  an incredible 70% of organizations had suffered at least one project failure in the prior 12 months!

A survey by Logica in 2008 showed that 35% of organizations abandoned a major project in the last 3 years.

A survey by KPMG in 2005 showed that only 2% of organisations said that their projects that year met all of their objectives.

Fixing Faults

According to Dosani 80% of technology projects cost more than they return because the benefits are overestimated and the costs underestimated.

According to the National Institute of Standards & Technology in the USA €60bn a year is spent on fixing faults in IT systems. Indeed 80% of the cost of a system is in making fixes to it.

In January 2013 it was announced that the Ministry of Defence had overrun by £468m on 16 projects in the past year with a total slippage of 139 months in the past year.

So, they have 16 projects that have overspent by an average of £29m each and they have slipped by an average of 8.7 monthseach   in the last 12 months.

There are some astonishing figures here that show that the world has not come to terms with successfully implementing software projects on a regular basis.

Just try getting the system you are going to develop insured against failure.

Change Needed

All over the world technology projects are failing and costing companies and countries incredible amounts of money.

Something has to change.

Companies need to take another look at the way they develop software. The whole paradigm is wrong.

Successful IT Projects

At one company where we ran IT projects for major companies like Texaco and Exxon (Esso) we managed to get 4 out of 5 projects done on time, with one of them well ahead of budget and ahead of schedule.

The only one that failed to deliver was one that we had to outsource as we didn’t have the right skills inhouse to do it.

Our people looked askance at the poor project management practices on the project as if they were looking at a backward people using backward methods and were not surprised when the project failed.

Huge Opportunity

Because there is such a huge problem there is also a huge opportunity. We intend to explain what needs to be done to get projects completed not just on time, to budget and fit-for-purpose but to be finished ahead of schedule and ahead of budget.

We change completely the way projects are run and we’ll explain how it is done in coming articles on this website.

In 2009, Roger Sessions, a well known author on the subject, in a White Paper called The IT Complexity Crisis: Danger and Opportunity, estimated that failures in IT projects cost a staggering $6.2 trillion a year globally.

That’s more than the GDP of the UK ($2.43 trillion) and Germany ($3.57 trillion) combined.

Houston, I think we have a problem!

I think we also have a solution!

Why Projects Overrun – 11 Reasons why Projects Lose Money

Why Projects Overrun

Most Projects lose money. Here we want to show you why projects overrun both time and budget, the main areas where money leaks away and how to stop the haemorraghing of profits from the project or the overrun of budget.

A Finance Director of my acquaintance once said, when I gave him the Project estimate plus or minus 30%, “You can forget the -30%. Projects are never under. You just hope that they are not too much over the +30%” – and he was, of course, correct to say that (but not on this project).

There are many reasons for this.

1.  Firstly, project estimates are usually wildly out because there is no historic data on how quickly an organisation can produce software and so the estimation process is completely flawed. If you were asked how much time someone would take to run a marathon without knowing what time they took previously your estimate could be wildly out. The first task before estimation should be to find out the previous speed of delivery of functionality which can be measured in Function Points per Man Month.

2.  Secondly tracking a project is wildly out mainly because incomplete tasks are counted in the tallies. How does one know how many errors there are left to fix in a piece of software? Developers underestimate the time left to complete to take the pressure off themselves. Who wouldn’t? Only completed tasks should be counted.

3.  Thirdly, there is no motivation for anyone to finish ahead of schedule. That means that the different components of a project are either finished on schedule or behind schedule – and unless you are very lucky and all the many components are finished on time then the project will be over time and over budget. There are several ways to cure this problem.

4.  Time is the most important thing on a project and that is what is measured and monitored. However, Fitness-for-Purpose and Quality are usually more important to the customer but these get sacrificed because Time is the most important driver whilst the project is running and the others are not usually measurable or measured during the project. The goals of the project should be aligned with what the customer wants and we have ways to measure this.

5.  They say be careful what you measure as that is what you will surely get. They also say Time is Money. However, this is not the case on projects. Because Time is all important, quality and fitness-for-purpose of the system suffers – and the rework to fix this scuppers the Time (and money) as well. This is the main reason for the failure of virtually every project and it is very curable.

6.  The Customer is King so they say – and who would want to upset a customer? Unfortunately, the customer, or those working for the main customer, are the ones most likely to screw up the project budget by failing to deliver information, equipment and data on time unless there are systems in place to make sure they can’t – and there seldom is. No one likes to kick the customer but there are times when the customer becomes the supplier and processes must be put in place to make sure they, like any other supplier, deliver on time – or else.

7.  External suppliers can also screw up the schedule, budget and profitability of a project by not delivering on time and there are seldom mechanisms in place to make sure that this is monitored and everything is delivered on time or there are severe penalties.

8.  The way projects are measured and the way those working on them are monitored is penal, de-motivating and bad for morale. There are better ways of motivating those working on projects and making them strive for success rather than having all their concentration on not failing.

9.  The person running the project may well have got to that position by being a good analyst or developer. However, the skillset for running a project is totally different from the skills and abilities they have shown so far. Any aptitude for their new role would be just a fluke. The project management role on a project should be split the way it is in the film business with a Director responsible for creating the system and a Producer responsible for the logistical side. This is currently done nowhere and brilliant analysts and developers are failing and causing major losses to their companies because they can’t estimate, track and manage suppliers and the other logistics that are needed to run a project successfully – skills which are nothing to do with being a good Analyst or Developer.

10.  The same mistakes are made over and over in projects as there is never time to look at what went wrong and how it could be put right with both a short term and long term solution. Even football teams take time on a Monday morning to go through the video to show how things went wrong so that the team can see how they can stop those things going wrong in the future. The same system would work on IT projects.

11.  One of the main ways that profitability leaks away from a project is through poor Change Control.  If this is not rigidly adhered to the customer gets a lot more functionality than they paid for. Of course functionality will grow as one looks more closely at a project. However, that is an opportunity for a supplier to make more money from a project rather than get the same amount of money for a greater amount of functionality. This is one of the main money leakers on a project. It should be one of the great money makers on a project.

Conclusion

Capitalism has been a much greater success than communism – yet at one of the bastions of capitalism, free market companies, their IT projects are run as if they are the command and control economies of communist states. There is little motivation for the workers to achieve. It would be much better to open the doors to capitalism and unleash the people on IT projects to achieve and to make the reward of success the main goal of people management on projects rather than the fear of failure.

The FD said to me “You can forget the -30%. Projects are never under. You just hope that they are not too much over the +30%”. That is certainly true in the way the great mass of projects are run. However, it doesn’t have to be so. We have shown you why projects overrun. However, there is another way.

Accurate Project Tracking for Successful Projects

Accurate Project Tracking

One of the most important factors in a successful project is accurate Project Tracking. Projects are seldom tracked accurately for reasons we will describe below. Bad data means bad information provided to management who then take bad decisions based on that bad information.

Time Pressure

Virtually all the pressure put on those working on an IT Project is Time Pressure. To everyone involved in a project being on schedule is the most important aspect of the project while it is being built. That is because time is money and it is easily measurable.

From developer to project leader to Project Manager to the Project Sponsors and senior people in the company time against schedule is by far the most important measure of a project when it is being built.

Of course, when a system is implemented and put into production there are other factors that become more important, i.e. where the system does as it is supposed to do, whether or not it is easy to use, the quality of the system in terms of how it works and how often or little it falls over.

These will all become more important eventually than time. However, they are less easily measured than time and are less important in terms of the tracking of the system during the build.

Huge Mistake

Virtually all the Project Management tracking during a project is about tracking time. That is a huge mistake – and counterproductive. It is the main reason that projects are late and over budget.

Be careful what you measure because that is what you get. There is tremendous pressure on developers to deliver on time.

So they do. It takes the pressure off of them. However, sometimes they would be better to take a little longer to test the software to make sure that it is working properly before handing it over for testing.

However, at that stage, no one is putting quality or fit-for-purpose pressure on them.  So what you get is software on time – but not necessarily fit-for-purpose or properly tested.

Early is Cheap

It has been shown by studies that the earlier you catch a problem in the software development cycle the cheaper it is to fix. If it is caught early in the cycle it is very cheap to fix. If it is caught late, e.g. in User Acceptance testing or in Production it is very expensive to fix.

And yet there is so little pressure during the software development process to get things right from te start. It’s all time, time, time. That’s what gets measured and this where all the pressure comes from. And yet time is seldom the most critical factor on a project.

Solution

So, how do we get round this? How do we make the most important factors on a project the most important factors while the software is being built? How do we pit Fit-For-Purpose and Quality pressure on developers during a project?

That’s very easy. You ask the customers what is important to them, how important it is and then you reward all those along the software developer supply chain for giving the users what they want.

Simple, isn’t it?

How?

So how do you do that?

Very easy indeed!

You just ask the customers or users. You just ask them what they want from the Project and how important it is to them.

They usually come up with Fitness-for-Purpose(it does what they asked for and what they want), Quality (in terms of not falling over), Ease of Use, Timeliness and Documentation. You then ask them how important these different factors are. You ask them to give a relative weighting factor.

Results

It will usually go something like

5 – Fitness-for Purpose

3 – Quality

2 – Usability

2 – Timeliness

1 – Documentation

There you are. You see how relatively unimportant timeliness is in the scheme of things. And yet that is the main thing that is measured during a project. That is the source of the main pressure.

Counterproductive

However, this is counterproductive. The more time pressure you put on the less Quality, Usability and Fitness-for-Purpose you will get – and that will have a huge impact on timeliness in the end as all of this will have to be fixed late in the cycle rather than early when it would have been much cheaper to fix.

Once you know what the user wants from your software development and from the system you are going to deliver to them then you have got to make sure that this is what is delivered to them.

Supply Chain

So, how do you do that? You create a Supply Chain from developer to analyst to systems tester to User Acceptance tester to the End Customer. At each stage whoever is receiving the software will give it marks for Fitness-for-Purpose, Quality, Usabilty, Timeliness and Documentation. They will be marks out of ten.

These marks will then be multiplied by the weighting factors so that the Timeliness mark will multiplied by 2 and the Fitness-for-Purpose mark will be multiplied by 5. These marks will be added together to produce a total mark for the delivery.

Best Return on Investment

If I was allow to introduce just one process at any company I went into, it would be this one. This one makes the most difference. The reason is that it totally changes behaviour.

Whereas previously developers would succumb to the time pressures on them and hand over work they weren’t completely satisfied with, I found that the developers and analysts themselves would resist the time pressures a little more as the realised that the bigger marks were for Fitness-for-Purpose and Quality and wanted to make sure that they got good marks for those.

They would hold on to their software for an extra day or two beyond schedule to make sure they were completely satisfied with it and it wouldn’t keep bouncing back to them.

Right First Time

Getting things right first time makes a huge saving in timeliness as it saved in the usual heaps of rework that timeliness pressure usually means so, you see, the less time pressure you put on and the more quality pressure you put on the more time you save.

What I found when I implemented this was that projects and people needed a lot less managing. People know what is wanted and managed themselves. In fact even when you put time pressure on them they’ll resist you as they take the long term view and deliver what is most wanted and what gets them the highest marks.

The easiest way to run a project is to put in good processes that reward people for the right things. That’s what you are doing here. Yet, how few project do this.