Why BIM Estimates Slip, and the Method We Use Instead

The most common reason a BIM project runs late is not a hard task. It is optimism. You look at something and decide it will take a day, and it takes three. Do that across a whole project, and the schedule slips, the team ends up working under pressure to claw the time back, and the client is left wondering why the date moved.

The fix is not to work faster. It is to estimate honestly in the first place. The method we use for that is three-point estimation.

How three-point estimation works

Three-point estimation, sometimes called PERT, replaces a single guess with three. For each task, you estimate an optimistic time if everything goes smoothly, a most likely time for the normal case, and a pessimistic time if it gets complicated. You then combine them into a weighted expected time:

ET = (O + 4M + P) / 6

The most likely case carries most of the weight, so the estimate stays anchored to the normal scenario. But the pessimistic case sits in the formula too, which pulls the number toward reality instead of toward the best day you can imagine. That single adjustment is what keeps the optimism in check.

A worked example

Take a Scan to BIM job on a 5,000 m² industrial building at LOD 300. This is simplified, since in practice we estimate in hours and split the work into far more tasks, but the logic holds at any scale.

• Point cloud cleanup. One day if it goes smoothly, two most likely, four if the data fights back. Expected time: 2.2 days.
• Architectural modelling. Five days optimistic, seven most likely, eleven if it gets complicated. Expected time: 7.3 days.
• The job as a whole. Six days at best, nine most likely, fifteen at worst. Expected time: 9.5 days.

A gut estimate would land on "about ten days" and stop there. The structured version arrives at a similar expected figure, around nine and a half days, but it carries information the gut number does not. The same job could plausibly run as fast as six days or as slow as fifteen, with nine the most likely. A flat promise of ten days hides the fact that a bad run is fifteen. With the breakdown, that risk is visible before anyone commits to a date.

Why it works

For the client, the estimate reads as logic rather than a number pulled from the air, which makes it both easier to trust and easier to question. Because the worst case is already priced in, a task that turns difficult does not blow up the schedule, since the time was accounted for from the start. The breakdown also makes resourcing concrete, because we can see which tasks carry the most risk and plan who works on them and when. For a discipline built on precision, the estimate deserves the same standard as the model.

Why "price per square metre" breaks down

This is also why estimating by area alone falls apart. A flat rate of, say, a thousand dollars per hundred square metres assumes every hundred square metres is the same amount of work. It is not. A hundred square metres of an industrial building, dense with structure and MEP, is a very different job from a hundred square metres of a hotel corridor. Pricing both the same way means underquoting one and missing the deadline on the other. The work drives the estimate, not the floor area.

The honest version

Three-point estimation does not make a project faster. What it does is make the estimate honest, which is usually what protects both the schedule and the client relationship. If you are tired of dates that slip and numbers that come from guesswork, the method is simple enough to adopt tomorrow. If you would rather hand the work to a team that already estimates this way, that is how we run every project.

See how we scope Scan to BIM work, or tell us about your project, and we will put a realistic number on it.