By the Numbers: How Setpoint Stays in Control
What lies at the heart of Setpoint's management system is a particular definition of, and way of dealing with, gross profit. In standard accounting, gross profit is calculated by subtracting direct costs from sales. The direct costs typically include all those expenditures for labor, overhead, and--in a manufacturing company--material that go directly into producing whatever is being sold. Hence the sum of those expenditures is called the Cost of Goods Sold, or COGS, and it rises or falls depending on the sales volume of goods being produced or services being rendered.
But did that method of calculating gross profit make sense from a project-management standpoint? Joe Cornwell, cofounder of Setpoint, a custom-manufacturing company based in Ogden, Utah, didn't think so. When you followed the convention, he reasoned, you implicitly assumed that direct labor and direct overhead fluctuated with sales volume, which Cornwell didn't buy. Regardless of the amount of business you had coming in, you still had to pay your full-time people, your electricity bill, your rent, and so on.
So wouldn't it be a more accurate reflection of reality, Cornwell wondered, for a project-based company to include only material costs in COGS? You could then view gross profit as a sort of aggregate income out of which you would pay all of your regular, ongoing expenses, including wages and salaries for regular employees.
If you defined gross profit that way, moreover, you could use it to monitor the efficiency of your operations. Here the concept of throughput comes into play. You can think of a project's gross profit (using Cornwell's definition) as a measure of its throughput--that is, the rate at which projects are completed using the company's resources. The more efficiently the system completes projects, the greater will be the benefit to the company. Indeed the company will benefit in two ways. It will earn more profit on the profitable projects, and the time not spent on a money-losing project can be used to work on a profitable one.
By the same token, you could use this particular definition of gross profit (from here on, we'll use "GP" to refer to it) to track the progress you were making on each project as you went along. Let's say you had a fixed-price contract on a project--that is, you'd negotiated a price to do the whole job. Going in, you'd know the expected revenues and cost of materials, so you could easily calculate the project's anticipated GP.
The next step would be to track the actual number of hours spent working on the project during the course of a week. At the end of each week, you'd add up those hours and, at the same time, you, or your project manager, would try to make a realistic assessment of how many more hours were required to complete the project. Suppose you figured you were 75% done, and suppose further that, at the end of the previous week, you'd decided you were 60% done. In other words, you'd completed 15% of the project--and thus earned 15% of the project's anticipated GP--during the week just ended.
That's an interesting number in its own right, but you wouldn't stop there. Since you know how many hours were spent on the project last week, you could also calculate how much GP was earned per hour. Then you could compare that number to the GP per hour of other projects and start asking the appropriate questions. Why were you earning more GP per hour on one project than you were on another one? Was the project manager doing something wrong? Was there a problem with the initial bid? Had unexpected problems cropped up? Should you try to get more money out of the customer?
Cornwell's definition also suggested a way of monitoring overall profitability. You'd begin by looking at what you were spending regularly on everything other than materials. Call those your operating expenses (OE), and include all the non-variable expenses involved in running the company, but not interest or taxes. Wages and salaries for regular employees would be included in those expenses, because Cornwell considered them non-variable. Contract labor, on the other hand, is truly a variable. So if contract labor were hired to work on a project, it would be added back into COGS, which means subtracting it from the project's GP.
GP is the amount of income you have available to pay those non-variable expenses. You can calculate the total GP earned each week by adding up the GP on all the projects. To get your weekly OE, you can forecast your monthly OE (we'll use that term because this is Setpoint's particular definition of operating expenses), divide by the number of working days in the month, and multiply by the number of working days in the week. When the GP earned last week equals last week's OE, the ratio of GP to OE will be 1.0, and you'll know you broke even on an operating basis. If the ratio is greater than 1.0, you made money; if it's less, you ran at a loss.
Those were the basic management accounting tools that Cornwell, with the help of Setpoint CEO and cofounder Joe Knight and cofounder Joe VanDenBerghe, came up with: GP, project GP per week, GP per hour, and the ratio of GP to OE. They were tools, moreover, that could be applied to all types of projects, including the time-and-materials variety, wherein the contract stipulates a certain hourly rate and materials markup but no total price for the job.
While Setpoint's project-management system works almost flawlessly on projects of 1,000 hours or fewer, the Joes found that it doesn't do so well on longer-term, more complex projects, such as designing and building a roller coaster. (Similar issues would arise with, say, software development projects, or the construction of apartment complexes.) The basic problem has to do with assessing how much of the project has been completed at any particular point. When there are too many labor hours involved in any one aspect of the project--such as mechanical engineering, for example, or design--the project manager simply can't make a reasonable judgment each week as to the number of hours left to be completed on that aspect. Any opinion is really just a guess.
One solution is to break the project down into a series of mini-projects of shorter duration, but that approach has problems, too. For one thing, it's difficult, if not impossible, to assign a meaningful dollar value to each mini-project.
The Joes have been looking for an alternative method, and they've come up with one that they're in the process of applying to the roller-coaster business. The approach takes advantage of project-management software, such as Microsoft Project, which allows you to break any project down into a finite number of tasks. Building a roller coaster might consist of as many as 300 separate tasks, for example. (Other companies have projects in which the number of tasks run into the thousands.)
If a project has a negotiated price, a finite number of tasks, and a projected GP (expected revenues minus material costs), it stands to reason that each task has a value. What's more, you can estimate the value of each task relative to the project as a whole and assign the task a specific amount of the project's GP. Once you've broken down the GP by tasks, you can create what the Joes call an "earned-value table," in which each task is listed along with its GP value and the number of hours the task is expected to take.
Using the earned-value table, it's a relatively simple matter for project managers to estimate how much GP has been earned at any point. They just look at the percentage of work completed on each task and multiply that by the task's assigned GP to get the earned GP. They then add up the earned GP on all the tasks, which gives them the earned GP on the total project.
Financial reports to outsiders
It's important to note, by the way, that Setpoint uses its unconventional definitions of GP and OE for management purposes only. Joe Knight adds direct labor and overhead back into COGS when preparing Setpoint's statements for its banker and CPA. Those people need financial reports based on Generally Accepted Accounting Principles to carry our their respective responsibilities. What the Joes need are tools for running a business, which demands a different way of thinking about the numbers.