Projects

Beating project deadlines without deadline pressure

Mohit Saini · 9 Nov 2025 · 4 min read

Beating project deadlines without deadline pressure

Look back at almost any delayed project, and hindsight makes it seem obvious: so many tasks could have started earlier. With more time, negotiations with suppliers would have gone better, alternative sources could have been found, and scope would have been better controlled.

So the lesson carried into the next project is: start everything as soon as possible.

Does starting early really make a project finish early? Usually not, and it often makes things worse.

Most tasks don’t decide the finish date

In a typical project, the large majority of tasks aren’t time-consuming enough to affect the overall duration. Starting them early doesn’t shorten the project. But can they delay it? Yes, and here is how.

The design engineer at rush hour

Take a task that appears in almost every large project: designing and releasing drawings. Civil construction, equipment manufacture, purchasing and many service contracts all depend on it.

The design engineer is usually working on many such tasks, often across several projects. Each needs inputs from other departments, outside agencies, suppliers and the client, and sometimes decisions from top management.

As the project moves into its visible, expensive phase (civil work on site, manufacturing of equipment), pressure on the designer peaks. People from every downstream team want drawings released so they can start.

Here is the key point: for an experienced engineer, making the drawing isn’t what takes the time. Gathering and integrating all the inputs is. When client inputs or decisions are missing, the engineer moves from one design to another, finishing none. To keep up credibility and relieve the pressure, the engineer does one of three things:

  1. Releases partial drawings so site work can start.
  2. Designs on assumptions, releasing drawings with conditions and gaps.
  3. Releases the easy drawings first: those that need little integration. The drawings on the longest path are usually the hardest.

Each has consequences:

  • With options 1 and 2, civil work slows down or stops midway. To keep civil crews busy, more work fronts are opened, which brings daily expediting, constant priority changes, multitasking, more interruptions and rework. Drawings go through many revisions, so the gap between drawings and actual site work never closes. (No wonder “as-built drawings” are a standard industry practice.)
  • With option 3, construction, manufacturing and purchasing get the easy designs early, and work on items needed much later in the project. Budget is consumed by work that wasn’t needed yet, and cash flow goes haywire.

In such an environment, expecting tasks to finish as planned is like expecting to win the lottery every week.

Integration points set the pace

Project tasks aren’t like production operations. Their durations are uncertain, and the uncertainty isn’t spread evenly. It is highest at integration points: tasks that need inputs from many others, like design release, final assembly and testing, or erection and commissioning. The more inputs a task needs, the less likely it is to finish on time.

The arithmetic is sobering. Suppose each input to a task has a 90% chance of arriving on time, which most managers would call reliable. A task needing two inputs has an 81% chance of having both on time. With five inputs, it drops to about 59%. With ten, about 35%. Design release, final assembly and commissioning often need many more than ten. That is why they are where projects get stuck, however well each individual input is managed.

When the project reaches these points, flow becomes turbulent and unpredictable. Pushing all tasks equally, without recognising these points, only adds to the turbulence, and makes it look as if bad planning is to blame.

A simple, robust approach

1. Identify the integration points. Before relying on any plan, find the tasks that need the most inputs to come together. We usually rank tasks by cost, duration or risk, rarely by how much synchronisation they need. Start ranking them that way.

2. Put a full-kit gate after them. Don’t let the integration task release its output piecemeal, and don’t let the next stage start until it is complete. For example, no civil work starts in an area until its full set of drawings is ready. This makes any delay visible immediately, which brings management attention at the right time.

3. Hold daily flow meetings. A gate exposes delays, but if delays aren’t prevented, the pressure to break the gate returns. Delays grow when information loops back and forth between departments, which no project network shows well. A short daily cross-functional meeting, chaired by someone empowered to decide, breaks those loops.

4. Plan with buffers. A large plan full of task milestones hides safety in every task and delays detection of problems. Move the safety to the end of each path as a buffer, where it is pooled and protects the whole. (Budgets aren’t handed out up front to each task owner; why treat time differently?) Release work in each area based on its full kit.

5. Prioritise by buffer status. In a large project, many tasks run at once, and not every delay matters to completion. Buffer management shows which ones do. A simple fever chart gives everyone a picture of every area. Combined with daily flow meetings on the ground and periodic buffer reviews by top management, it does far more for the project than chasing dates.

Projects aren’t won by starting everything early. They are won by getting through the integration points smoothly.

Related reading

Mohit Saini

Strategy, supply chain & sales transformation consultant. About · Book a discussion