57 lines
2.5 KiB
Markdown
57 lines
2.5 KiB
Markdown
---
|
|
id: 2026-01-11T11:00:00-0500
|
|
title: 2026-01-11 11:00:??
|
|
tags:
|
|
- topic/estimating
|
|
daily: "[[2026-01-11]]"
|
|
---
|
|
# 2026-01-11 11:00:??
|
|
|
|
![[my-favorite-quotes#"It Takes an Engineer to Build a Bridge that Barely Stands"]]
|
|
|
|
The value that estimators provide for a contractor
|
|
is in modeling project cost.
|
|
If the goal were simply to present a number the cost will not exceed,
|
|
than anyone could be an estimator.
|
|
If the goal were only to present a reasonably accurate figure,
|
|
with no other constraints,
|
|
then there are few who in the world who couldn't,
|
|
given infinite time for a single bid.
|
|
The _true_ value of an estimator, then,
|
|
is in their ability to model project cost _efficiently_, that is,
|
|
to achieve _acceptable_ accuracy and precision as quickly as possible,---
|
|
much as the engineer's is in building a bridge that _meets_ the requirements
|
|
as cheaply as possible.
|
|
|
|
Estimating "as accurately as possible"
|
|
is akin to building a bridge "as strong as possible";
|
|
it sounds nice, but ignores the actual objective of _optimal cost-efficiency_.
|
|
|
|
I think that most estimators believe this to be the case,
|
|
but, whatever they believe in mind, they believe in practice
|
|
that the goal is to model cost as accurately and precisely as possible,
|
|
_given the time allowed for bid_.
|
|
Estimates tend always to take exactly as many weeks as until the bid due date,
|
|
even those of significantly different turnaround but equal scope and complexity.
|
|
This behavior is objectionable,
|
|
since if an acceptable estimate _could_ be provided in two weeks,
|
|
it is not cost-effective[^1] to allow it to take four.[^2]
|
|
|
|
[^1]: "Cost" referring both to estimator salaries and the opportunity cost of declined bids.
|
|
|
|
[^2]: Supposing a contractor maintained a [[modern-portfolio-theory]] styled record of estimates
|
|
including pending bids, and projects ongoing and completed
|
|
(each with confidence estimates),
|
|
uncorrupted by [[estimating-culture#Incentives|perverse incentives]],
|
|
they may have reasonable basis to set sliding standards for estimate precision
|
|
to be specifically determined at consideration of the opportunity for bid
|
|
according to current climate (i.e. their transient risk appetite).
|
|
I believe my use of the absolute is still fair:
|
|
No contractor is doing that, so they lack a competent measure of risk tolerance
|
|
besides continuing to tolerate what they have historically.
|
|
|
|
> [!important]
|
|
> Objectionable as it is,
|
|
> it is to be expected that estimators will use all time allowed them
|
|
> when the standards for _acceptability_ are inadequately defined.
|