How can I stop a runaway project failing?

One of our IT projects seems to be spinning out of control. The project manager, however, denies the existence of the problems....

The question:

One of our IT projects seems to be spinning out of control. The project manager, however, denies the existence of the problems. Given that this project is viewed as critical by the business, how can I ensure that it does not end in failure?

The answer

Proper reporting could have prevented this problem

Robin Laidlaw, president, CW500 Club

"Seems to be spinning out of control" is a strange phrase to use. There are a number of issues to be considered, some managerial and some technical.

Do you have a project reporting methodology? It seems that you do not, otherwise you would not use that strange phrase. The reports should indicate if the project is running late, over budget, has been subject to lots of (unauthorised?) specification changes. If it is showing red alerts on all such aspects then yes, it is out of control - but you should know that, not just be concerned that it may be.

By what criteria is the project manager saying it is OK? He must have data - why not show it to you? If he will not show you his project progress reports then you can be pretty sure there is a problem. If he does not have such reports then you are running a rather slack ship.

Project managers are usually given projects to manage so, if this project is beyond his capability, then it is the fault of the person (you?) who gave him the assignment. If he was given a project that is beyond him, and it is going wrong, then it is not surprising if he is creating some smoke-screens - our industry is littered with past project managers who have been fired for project performance problems.

You say the project is mission-critical: you must therefore act immediately and decisively. You must instigate an urgent review of the project - maybe bringing in an external group who will give you an unbiased report - and then take the action needed to get the project as back on course as soon as possible. Report to the board immediately; if they are planning revenue and profit dependent on this project's original timescales, costs and functionality, the longer they are kept in the dark, the more angry they will be.

When you have recovered the project as well as you can, then you must review the managerial processes which led to this situation which appear to be threefold:

  • Who signed off the project for board approval with this project manager in charge?

  • Who gave the assignment to him, if he has not convincingly proved in the past that he could handle it?

  • Why have you not got proper project progress reporting systems in place which automatically send reports to the project manager's boss?

If the answer to one or all of the above is down to you, then you have put yourself in a vulnerable position.

Consider whether it is time to walk away from the project

Ollie Ross, head of research, The Corporate IT Forum, Tif

I am presuming that if this project is critical to the business it has board-level ownership; that it was approved using a structured process to outline the deliverables, benefits, and metrics for measuring success against these criteria; that milestones and key stages were agreed and are reviewed using an approved methodology; that informal and formal communication between all involved, responsible and accountable occurs regularly; and that slippage issues or irregularities are identified early and dealt with collectivelyÉ yes?

Then you should have the evidence to hand that will either support your project manager's confidence, or justify your fears.

But if facts and figures are not available and there is no means of collecting them, the board and the business is as much to blame as the project team and all must now work together to rescue a spiralling situation.

Is there a clear finish line? Stick to it. The project must have a clean end point. Is it a question of scope-creep? Refocus on the core deliverables and cut the frills and secondary initiatives. Are there problems with suppliers or the technologies? Could you walk away if you needed to?

Fast-changing or turbulent market conditions regularly give rise to problems in the time it takes to deliver some projects. Business benefit cases either do not materialise as expected, or can disappear overnight. Reduce these risks by setting up regular "gateways" at each phase of the project where the business reviews market conditions (and whether the money being spent on the project is still worth it) and decides if the project should continue.

It is vital to keep overall control of a project

David Threlfall, director of MIS, Enesco

As project "sponsor" it is vital that executive IT always keeps overall project control without undermining the project manager.

The perception by departments outside IT as to the success or otherwise of each project may not reflect how the project is progressing on the Gantt chart, however, perceived success is how company confidence in IT projects is developed and will affect support for future proposals.

To avoid poor perceptions, you need to:

  • Have a clear and realistic definition of project goals

  • Ensure full company buy-in; involve all departments, from user to executive, that can be affected by the project, no matter how minimal

  • Define clear, realistic milestones

  • Recognise any personality conflicts and address these quickly and professionally

  • Avoid a blame culture, learn from mistakes, support team members and move on

  • Most importantly of all: communicate.

Appoint a mentor to help with risk management

Sharm Manwani, Henley Management College

Any successful IT project will be managed "hard" and "soft". The former relates to having the right project environment, including methodology, tools, and reporting. On the softer side, you are clearly one of the priority stakeholders whose expectations are not being managed.

From what you have said, the project manager is not managing risk effectively. You do not specify what structure you have in place to do this, but you can certainly influence the development and maintenance of an effective risk register based on probability and impact. This will ensure a shared understanding of risk and might actually allay many of your concerns. Ideally in this business-critical project, you would have a contingency option where you can take personal action.

Project managers are successful because they have a strong focus on delivery and are happy to take full responsibility. You need to find the right balance between maintaining that personal motivation and being close enough to take action if needed. Consider appointing or acting yourself as a mentor. Explain how easy it is to go from "hero to zero" as a project manager, and the importance of both methodologies and stakeholder management.

Consider bringing in a heavy-hitter to rescue the project

Chris Potts, director, Dominic Barrow

It sounds like time to bring the full force of your company's people-management strategy, sourcing strategy and risk policies to bear on the situation.

The obvious option is to replace the project manager with someone else. I assume the situation is more complex than this or you would not be asking the question. If, for example, you have no one else on your staff who could manage the project in his place, then bring in an external heavy-hitter on contract who will also strengthen overall business project management. You will have to pay more in the short-term but the return on investment will be worthwhile.

Alternatively, I know there are sometimes political considerations that make such straightforward solutions difficult. One "politically correct" option is to manage the problem via one of the suppliers involved in the project, although this is risky and not usually the best alternative for many reasons.

It is time your project had a thorough healthcheck

Gill Williams, partner in Ernst & Young's information security practice

Some people work best when up against a deadline, either because they need deadlines to motivate them or so that they can always come through and "save the day".

Most organisations will not tolerate this approach to project management. It is not only about delivery against scope, budget and timescale but also about business success, which means getting stakeholder buy-in.

Your feeling that this project is "spinning out of control" may be because you are starting to see some red flags such as loss of project pace, out of date plans, vague progress reports; or because you do not have sufficient visibility about what is happening. The key activity now is to get that visibility, particularly of progress against plan, the status of change management and resourcing, and the risks and issues status.

A "healthcheck" can be done at any point in a project and it does not need to be time-consuming. It is never too late to get a clear view of what is working and what is not. It may be helpful to get that view from an independent party, either from elsewhere in your organisation or externally, to get a truly objective view.

Do not be put off if the project manager says he is too busy to take part in a review - it is better to take some time now to be confident that the project will deliver, or to be able to take steps to get it back, on track than to wait for it to fail. Remember that it is the project that is under review, not its manager, (although resourcing and capability of the steering group, the project sponsor, project manager and the end-users should be part of the review). Finally, take this opportunity to put in place good project governance going forward for all projects.

Get close to the project manager and be supportive

Rob Lambert, senior lecturer in IS, Cranfield Information Systems Group

It would appear that you are facing two issues. First, at an organisational level you do not seem to have effective project planning and review processes in place that would apply as a matter of course to all projects. Therefore you need to review these processes to ensure there is clarity and agreement among your staff as to the purpose of project reviews, the documentation to be produced and the quality of the data to be discussed at the meeting.

Effective review meetings are meant to help both the organisation and the individual project manager, they are not "witch hunts", although this does seem to be the case in many blame cultures. Good review meetings should enable project managers to identify issues and concerns they have that may affect the outcome of the project .

Second, you need to understand why the project manager in question appears to be so closed and defensive. Is it that he perceives review meetings as bureaucratic and of little value? This you need to address by discussing the value, both to the organisation and the project manager, of effective review meetings.

Alternatively it could be the project is "spinning out of control" and you certainly should know about this. Hence it is imperative that you get close to this project manager and ascertain, in a supportive way, the reason for his apparent defensive behaviour.

You need to assess and then manage the risks

Roger Rawlinson, NCC Global

Your issue is risk. You say in the past that your project manager has always delivered, so one option is to trust him on the basis of his track record.

The problem with this approach is that you may have trouble sleeping at night and, if the project fails, you will have failed in governance and due diligence.

As this project is business-critical, the risks are even higher in terms of the impact of failure, so it is vital that you talk to your project manager. Discuss your requirements in terms of you reporting criteria to the business and the implications of this process.

Discuss your requirement for project metrics that you need to present upwards. It sounds as though this type of reporting will be new to the project? In which case you will need to be tactful in how this is introduced. You need to sell it as a benefit to your project manager; explain it will help them as well as you, it will enable them to escalate issues upwards and therefore manage expectations and risks.

The situation that you describe is not good. It may be that your project manager is saying everything is fine, but in reality he may be panicking and your style of "arm's length" management, leaving it to the project manager to deliver the goods, makes him reluctant to approach you with his problems. Sit down and talk to him about this in an informal environment.

Learn to handle personal relationships with all parties

Roger Marshall, Elite

You would be well advised to hope for the best but plan for the worst. Your judgement is probably right and the project manager's is wrong.

There is no substitute for getting involved in the detail yourself (or asking another experienced person to do so on your behalf). Insist that the project manager takes you through every line in the plan and explains why he is confident that it is in order. You will either end up convinced or with a set of action points that will involve extending the plan, adding resources or cutting the deliverables, but probably a combination of all three.

You will then have to sell the new plan to the customers, while getting all the grief over in one go. Naturally the new targets must be realistic and you will need to maintain your personal monitoring role - this is now your project as well as the project manager's. Handling the human relationships - with the project manager and the customers - is the hardest part.

The experts

Computer Weekly has put together a panel of experts. You can draw on their specialist knowledge to solve a problem. E-mail your questions (or your own solutions to this or the next question) to [email protected]   

NCC Group 

Deloitte &Touche 

Cranfield School of Management 

Computer Weekly 500 Club 

Henley Management College 

British Computer Society 


The Corporate IT Forum 

Dominic Barrow 

Ernst &Young

Read more on IT project management