Slack

Slack is the unplanned time and energy that we allocate for high value and high priority work.

Civil Engineers and Estiamting

When rounding numbers in "math," the common notion is to round to the 'nearest' whole number. To round down if .5 or under and to round up if higher than .5 (ie .51 or .501). For general math, this makes sense as we are trying to get rough numbers to represent the equation we are sorting.

In civil engineering however, the notion I have found is to always 'round numbers up' even if the decimal point is at the bottom half of the number. That means 12.8 rounds up to 13, but so does 12.2

If we are building a highway that runs 12.2 miles and we order materials and labor for 12 miles, rounding down - we will end the project short .2 miles (approximately, as the '12.2' number is likely a rounded number) of materials, and labor. Ordering 12 miles of concrete and rebar doesn't stretch to 12.2 miles very easily, and sourcing the materials and labor for the last .2 miles is likely to take just as long as the original 12 miles did.

On the other hand, if we are building a 12.2 mile highway and we order materials for 13 miles, rounding up, we might end the project with .8 mile of extra material or labor (we always need a little extra anyway), but the commitment is still completed based on the cost estimation provided.

Estimating a little short means missing the commitment, estimating a little long means we have additional material.

What civil engineers are doing here, and what we can do with our own estimates is creating slack.^[1]

Human estimating error

Humans are inherently poor at estimating time and effort commitment. It is a skill that we can improve upon, but the relationship between our estimate and our execution is very complex.

It is true that we often find work expands to fill the time provided AND we tend to expect things to move faster than they actually do. Both of these are true and yet, they have opposing stances on how we use and fill the time that we have.

Let's say we are planning out a 40 hour work week. If we have rough estimates for all of our work (including clerical and administrative work, as well as breaks) we will inherently underestimate some items and overestimate others. On the whole, it ought to average out so that we end up with a work week that is about 40 hours, sometimes more and sometimes less.

When the work we are doing is in the upper right quadrant of the Eisenhower Matrix (important but not urgent) we have the ability to fill up our commitments. We plan out 40 hours of effort, then if we end up with 38 hours of actual work- we have 2 hours to investigate the next thing, or learn, or rest or whatever. If we end up with 42 hours of actual work, we can pick it up next week. After all it is not urgent work, it can wait until next week.

When we are doing work in the upper left quadrant of the Eisenhower Matrix, (important and urgent) we want to build slack into our effort estimate. A few extra hours at the end of the week means the same thing, we can investigate, or learn, or rest. A few hours short means we missed the commitment, and when the work is urgent as well as important- we miss the opportunity- (We don't prefer to work nights and weekends).

Like in civil engineering, erroring on one side of the line is a lot more manageable than erroring on the other side.