Gantt chart vs PERT chart

People treat the Gantt chart and the PERT chart as competitors, often because both turn up in the same project-management course back to back. In truth they're two views of the same underlying plan, and they answer different questions. A PERT chart shows you the logic of a project — what depends on what. A Gantt chart shows you the schedule — when each thing happens on a calendar. Used together, the PERT diagram works out the structure and the Gantt chart presents it.

What a PERT chart is

PERT stands for Program Evaluation and Review Technique. It was developed by the US Navy in the late 1950s to manage the Polaris missile programme — a project so large and uncertain that no calendar drawn up front could be trusted. A PERT chart (more generally, a network diagram) draws a project as a set of boxes connected by arrows. In the common activity-on-node style, each box is a task and each arrow is a dependency pointing from a task to the one that follows it. Trace the arrows and you can see every path through the project from start to finish.

The defining feature — and the thing that trips people up — is that a PERT chart has no calendar. There's no time axis. A box doesn't get wider because a task takes longer; the diagram only shows order and dependency, not duration drawn to scale. That's the point: it strips away dates so you can reason about the structure of the work first, without the distraction of when anything lands.

What a Gantt chart adds

Take that same network of tasks and dependencies, give each task a duration, and place it on a calendar, and you have a Gantt chart. The Gantt chart adds the one thing PERT deliberately leaves out: time, drawn to scale. A long task is a long bar; you can see overlap, you can see how many weeks the whole thing runs, you can drop a today line on it and track progress. What a PERT chart expresses as pure logic, a Gantt chart expresses as elapsed calendar time.

This is exactly why they're complementary rather than rival. The PERT diagram is the better place to reason about structure — to ask "is this dependency real? can these two run in parallel? what's the simplest network that delivers this?" The Gantt chart is the better place to communicate and track the result, because dates are what stakeholders and calendars actually deal in.

The three-point estimate

PERT contributed one genuinely distinctive idea to estimation. Instead of asking for a single guess at how long a task takes, it asks for three:

• Optimistic (O) — if everything goes right.
• Most likely (M) — the realistic, normal-conditions estimate.
• Pessimistic (P) — if things go badly (but not catastrophically).

It then combines them into an expected time with a weighted average that leans heavily on the most-likely figure: (O + 4M + P) ÷ 6. So a task estimated at 2 days optimistic, 4 most likely, and 12 pessimistic comes out at (2 + 16 + 12) ÷ 6 = 5 days, not 4 — the long tail of the pessimistic case nudges the expectation upward, which is usually wise. The spread between O and P also gives a rough measure of how uncertain the estimate is, which a single number hides entirely.

This is a useful discipline well beyond PERT itself, and it connects directly to the wider craft of estimating task durations. Even if you never draw a network diagram, asking "what's the best, likely, and worst case?" produces more honest numbers than demanding a single confident figure.

How PERT relates to CPM

PERT is often mentioned in the same breath as CPM, the Critical Path Method, because the two were developed around the same time (the late 1950s) and do overlapping things on network diagrams. The honest summary is that they've largely merged in modern practice, but they started with different emphases:

• CPM assumed durations were known and fixed, and focused on finding the longest path through the network — the critical path — and on the trade-off between time and cost (how much it costs to speed a task up). It grew out of industrial and construction work where past data made durations predictable.
• PERT assumed durations were uncertain — hence the three-point estimate — and focused on probability: how likely is it that we finish by a given date? It grew out of research-and-development work where nothing had been done before.

Both find the critical path on a network of dependencies. The difference is just whether you feed that network single durations (CPM) or probabilistic ones (PERT). Today most tools and most people blur the two together and simply say "critical path analysis on a network diagram", which is fine.

When to reach for each

Reach for a PERT or network diagram early, when you're still working out the shape of a project — when the dependencies are tangled, when you want to find genuine opportunities for parallel work, or when durations are so uncertain that a calendar would be premature. It's a thinking tool for the planning phase.

Reach for a Gantt chart once the structure is settled and you need to schedule, communicate, and track — which is most of the time, and for the whole life of the project after planning. It's the working view you live in.

How modern tools blend them

The practical reason the "vs" framing has faded is that modern software collapses the distinction. You build a Gantt chart, draw the dependencies between tasks, and the tool computes the critical path for you and can show the network behind the bars — the PERT logic is running under the hood whether or not you ever look at a box-and-arrow view. The three-point estimate survives as an optional input where uncertainty is high. So the typical workflow today isn't "PERT or Gantt" at all: it's a single dependency network that you can view as a calendar when you want dates and as a network when you want to reason about logic. The two charts are just two windows onto the same plan.

Frequently asked questions

What is the main difference between a Gantt chart and a PERT chart?

A Gantt chart shows tasks on a calendar — durations and dates drawn to scale. A PERT chart shows tasks as boxes connected by dependency arrows, with no time axis at all. PERT shows the logic of the project; Gantt shows the schedule.

What is the PERT three-point estimate formula?

Expected time = (Optimistic + 4 × Most likely + Pessimistic) ÷ 6. The weighting of four on the most-likely value makes it the dominant term, while the optimistic and pessimistic bounds pull the result toward the more probable outcome and reveal how uncertain the estimate is.

Are PERT and the Critical Path Method (CPM) the same thing?

They are close cousins, developed around the same time, and have largely merged in practice. CPM originally assumed known, fixed durations and focused on time-cost trade-offs; PERT assumed uncertain durations and focused on probability. Both find the critical path through a network of dependencies.

Do I need a PERT chart if I already have a Gantt chart?

Often not as a separate artefact. Modern Gantt tools compute the critical path and hold the dependency network for you, so the PERT logic is already there. A standalone network diagram is most useful early, when you are still untangling dependencies and looking for parallel work.