There is no better way to test your assumptions than by standing in them.
Estimating accuracy does not come from a better rate library. It comes from knowing whether the rate applies to this building, on this site, on this street. Cost plans are built out of validated numbers, and validation happens where the work happens.
A cost plan is only as good as the assumptions under it, and assumptions never announce themselves as wrong. They sit in the model looking exactly like facts.
A guess is not a rate
Estimators are far too often throwing around guesstimates without understanding the implications.
With every guess, you should be validating your number. That is how great cost plans are built.
A rate is an average of jobs that were not this job. That makes it a good place to start and a poor place to stop.
Go to site. Do not walk, run.
What the site tells you that the drawings do not
Watch how difficult it is to move concrete from ground to level six, and you will realise your standard m3 rate does not apply here.
Watch structural steel being craned onto a level four rooftop in the middle of the CBD, and you will see the true cost of out of hours work and the difficulty of install.
Neither of those costs is drawn anywhere. Both of them get paid.
The refurbishment that proved it
We recently priced a heavily refurbished job for a builder, and won it. After walking the site, we confirmed what every experienced estimator already knows. Drawings rarely tell the full story.
That gap is widest on refurbishment work, where a large share of the job was never documented in the first place.
The gap between two allowances
One estimator might allow $50k for make good. Another might allow $75k. Later you find out you lost the job by the difference.
Overallowing and underallowing is where we all get caught out, and that is completely fine. The only way to balance that is through experience.
There is no lookup table for that gap. The difference between the two allowances is not a measurement error. It is a judgement about how much of the existing building is going to fight you, and you only get better at it by having been on the wrong side of it.
It still humbles me
You will trip over your feet a few times before you get it right. Even then, some days, you will still get it wrong.
That is what makes this game humbling. Still humbles me.
A site walk does not remove the risk from a cost plan. It moves the risk from something you guessed at to something you have stood in.