Whipping our behinds

The Egan Report’s target of an annual reduction of 10% in construction time is a modest aspiration, and certainly achievable. The received wisdom that time improvements are earned at the expense of cost and/or quality is incorrect.

The overwhelming evidence of the National Economic Development Office report Faster Building for Commerce in 1988, based on 8000 projects, was that, up to a point, the shorter the planned times were for a project, the greater the likelihood it would be finished on time. Why? The evidence suggested that a short time span keeps everyone focused on meeting the deadlines. Recent management books have developed the theme and, one in particular, EM Goldratt’s Critical Chain, suggests a practical way to maintain that focus in project planning.

It starts with those involved in the project deciding how long tasks take, a process that inevitably results in excessive “float” built into every task. On a recent project, the surveyor suggested that the programme should allow two weeks for the survey.

The problem …

At face value, this was not unreasonable, but on detailed examination it transpired that the surveyor would be on site for one day, would need another to process the results, and a further day for cross-checking, correcting and printing. He allowed himself a bit of extra time in case he did another survey that week and an additional margin in case the printing was held up. A hard look showed that every party to the programme had a similar approach. Everyone appeared to give “realistic” estimates based on their worst experiences, and a huge float was factored into every task.

Other human factors militate against brevity. Estimators add in float in the expectation that it may be cut. Managers protect their backs by adding float and, the more levels of management, the more float added.

Despite all this float, projects are still commonly late. Why?

• Time management is poor. Tasks tends to be finished late, blamed on pressure of work, other people, management, but rarely ourselves. Yet, in theory, if the first task were brought back in line and started on time, all subsequent ones would be on time. This never seems to happen.

• Everything is left to the last moment. If it is thought that the task would take a week and the deadline is in four weeks, it probably won’t be started until week three. The task will either be just in time or late, but never early.

• Too many tasks are attempted at once. The first of three one-week tasks could be completed at the end of week one, the second at the end of week two and the third at the end of week three. But it is usually the case that all three tasks are performed together and so all the tasks each take three weeks.

• Tasks that are started late are usually finished late. In most people’s minds, the time a task takes is what is important, rather than the dates. “I’m supposed to have a week to do this and I’m not going to kill myself to do it in three days just because they gave it to me late.” When tasks are linked, delays are rarely made up for.

• When a task is finished early, the final aspects of the task tend to be dragged out (“no need to get it typed up until Monday”), or it is sat on to prevent the person involved being loaded with another task.

Even if early completion is reported, the chances are that the next sequential task will not be able to start before its programmed date because of other commitments. Furthermore, those next in the chain will feel that there is now even more time to carry out their task and even less reason to rush. In the case of simultaneous tasks, it is always the biggest delay that is passed on to the next dependent task.

Everyone appeared to give “realistic” estimates based on their worst experiences, and a huge float was factored into every task

… and the solution

What can we do about all this? As the authors of the NEDO report discovered, the key to fast building is focus. “Slow building often signified lack of purpose or momentum, drift, under-resourcing, second-class attention to the project and a disappointing outcome all round”, yet “projects with fast planned times were also likely to finish on time or early”. When time is the focus, project performance follows.

So, how do we get the project team to focus? How do we get rid of the excess float in every task? By recognising that we have shot ourselves in the foot by setting milestones and durations in stone.

Goldratt’s key suggestion is that we strip out all individual task float but allow some project float after “completion”. Thus, completion becomes the project’s early finish date with a backstop of a late finish date. The likely completion date is somewhere in between.

Management and communication now become key. The project team needs a daily reminder of the progress of preceding tasks and the planned start date of their task. The result is that when a task is ready to be worked on, the team should start on it immediately, knowing that, in the time available, they only have, say, a 50% chance of completing it on time.

As everything is now critical, the project manager is able to chase progress on a daily basis and this daily prodding helps to overcome poor time management. To achieve the task, the team will have to drop everything and become more focused. This is a cultural change and will not come easily.