Estimating labor can be a daunting task; often, it seems to boil down to a nebulous, “we should be able to get this done in this many hours,” kind of statement, which turns out to be neither precise nor terribly accurate. This is because a number of variables have to be taken into account when considering labor: the particular skills and abilities of the employees under consideration, the complexity of the tasks to be estimated, competing tasks putting demands on available labor, bottlenecks in shop flow due to availability of resources and materials. Even considerations like climate control (consider a non-air-conditioned shop during that ridiculous heat wave two weeks ago!) and the morale and mood of your employees can have a major impact on the rate at and efficiency with which your build progresses.

At base, the process of estimating labor is no different than that for estimating material cost: there are three basic approaches (analogous, parametric, and bottom-up) which increase in accuracy as well as the effort required to complete them. With material estimates, however, we have ready access to some basic information—we can call and get prices for specific materials. Unfortunately, there’s no easy way to call up your local big-box home improvement store and ask, “how many hours will it take Johnny Carpenter to build this scenic wagon?” Additionally, we can sometimes save money by lumping some things together—a larger lumber order saves on multiple freight charges, for example, or we can save a couple of bucks per stick of lumber by ordering 1x6 instead of 1x4. The same isn’t true of labor, necessarily: putting two employees on one unit doesn’t necessarily cut the task time in half, and, depending on how well they work together, this may even lengthen the time required to complete the task!

Any approach to estimating starts with information, and it pays to spend some time gathering data about your particular shop. By keeping task-based time records, you can fairly quickly develop a database that can tell you how quickly different employees in your shop can complete different tasks. Asking employees to enter the particular tasks they have been working on and the time they spent on them may seem to be a time-waster, but there are ways to make the process less-impactful, and the payoff can be enormous; consider putting log sheets on clip boards at every workstation, within easy reach of your staff, or asking staff to keep a small pocket notebook with them in which to record notes related to time on task. A couple of scenic builds should provide enough data to get started; from this information we can start assessing how long particular projects take, as well as how employees compare in terms of efficiency with particular kinds of projects. You already know intuitively that Johnny isn’t nearly as quick an up-rigger as Billy is; with this data, you can start assessing how much slower Johnny is, and what impact that will have on a build that requires a lot of up-rigging.

Your goal with this information should be threefold: first, you should be able to assess quantitatively who is more efficient at what kind of jobs, aiding with the scheduling process; second, you should be able to create a reasonably accurate parametric model to aid in the estimating that typically happens in the midst of the design process; third, you should be able to create a table or list of data to use in the bottom-up estimating typical of the budget-approval process just prior to beginning a build. All of this data provides tangible evidence for your estimates that you can both rely on to feel confident in your numbers, and point to as backup for your assertions.

Unfortunately, unlike estimating material costs, estimating labor doesn’t stop with a tally of the number of labor hours required to complete the task. Scheduling a project provides a more accurate determination of when the project will be complete than simply tallying the number of person-hours required and comparing it to the number of person-hours available. (See How Long Will it Take, PERT Charts, Critical Path Method, and Calendars and Gantt Charts.) And, as discussed last week, unlike money, you can’t just throw more labor at an over budget build—overtime pay can be devastating to a company not prepared for it, and often, additional people are not readily available. Coming to the table with a bottom-up estimate coupled with a comprehensive schedule based on the critical-path method provides a clear picture of the labor impact of a project, and enables all of the stakeholders in the project to have a good understanding of the total cost of the project in terms of time and materials.

One last note: some venues look at labor as a function of time; others, because staffing is determined on a project-by-project basis, look at labor as a function of money (this is particularly true at a scenic shop, for example, that bids on a project based on a specific labor cost, and needs to stay within that estimate in order to make money or even break even on the project). Additionally, converting labor hours to labor cost is a relatively easy task, and can often cast a project in a slightly different perspective. Consider, for example, a recent production I was a part of: the scenic materials budget was only about $2,000—not a significantly large sum of money for a regional show, certainly. In terms of labor, however, we spent something like $14,000—including salary for the technical director, assistant technical director, two staff carpenters, and four overhire carpenters.

Though the build for this production was only three weeks, the amount of money spent on labor during those three weeks was fairly significant; additionally, even adding one overhire carpenter for a full week’s worth of work would raise that total by another $600 or more. This is important to keep in mind as you enter into negotiations about how much is accomplishable in the time allotted with the available resources—it can be easy to say, “call in some overhire,” but when you start tallying up $600 to $1,000 per person per week of labor, labor costs can spiral out of control—and those numbers carry more meaning, sometimes, than saying a project will require an additional 40, 80, or 120 hours of labor in a given week.