In a discussion on FaceBook regarding markup my friend and collegue Dan Kolbert of Kolbert Builders in Portland ME wrote:
“…the most predictable aspect of your volume is your own crew in the field, and your markup should be connected to that, rather than something much more variable, like your overall volume, the expense of materials, etc. Gerstel [David Gerstel Nail Your Numbers: A Path to Skilled Construction Estimating and Bidding] argues for the number of jobs your leads can run, figuring out how much OH/P needs to be recouped per week per lead, and divvying things up. Others (myself included) look at the overall estimated billable hours your crew will get in for the year, and making sure your jobs recoup OH/P based on those numbers.”
What he is talking about is illustrated in this chart. Back in January of 2009 I was trying to teach myself Apple’s Numbers program to see how it compared to Excel (to make a long story short it doesn’t compare, stick to Excel). So I had just gotten the 2008 financials from one of the contractors I was working with as a coach and consultant back then so I decided to see how charting in Numbers worked plotting out his Sales data for the year.
I noticed the chart I was working on perfectly illustrated what I have been writing about in regard to the problems in using a Volume Based Markup approach to pricing work. In a Volume Based approach the Total COGS can bear little relationship to the month to month real cash flow requirements of a company. In months where Material and SubContracting sales are lighter a company can be cash strapped to even cover their Overhead requirements. So I thought I’d publish the chart and see if everyone else would see what I see.
Another collegeue of ours Bob Kovacs at the time Vice President of Preconstruction- Skanska USA Building wrote:
“Hmmmm…I think we can gather that labor costs and overhead costs stay rather consistent from month to month, whereas the other costs vary greatly from month to month. Seems like a good reason to tie (allocate) overhead costs to labor costs, rather than to total volume.”
This was a small Connecticut based contractor and as I recall it was just 3 guys working together. They did general carpentry, mostly exterior, but a lot of custom decks in the summertime and during the colder winter months they worked inside picking up a lot of interior painting jobs to fill the gaps in their schedule so that is why the material costs went up in the summer months and down in the winter.
Back in 2004-05 I collected and compiled some data on different project types to see what the ratio of Materials to Labor Hrs was for different kinds of projects or more specifically how many dollars of materials were being installed per Labor Hr. concrete and masonry work was the lowest and kitchens were the highest but the message is/was not all work is the same in the ratio of time the project takes to the material cost involved.
So contractors I have talked to argue their market, their work product doesn’t vary seasonally but that ignores the fact that even if you do the exact same kind of project all years long this same kind of variation takes place within a certain type of project. Not all task have the same ratio of materials (or subs) to you own labor’s time.
Recently yet another colleague on FaceBook reminded me of a lesson learned in a book I read years ago (The Flaw of Averages: Why We Underestimate Risk in the Face of Uncertainty by Sam Savage) which teaches us: “…plans based on assumptions about average conditions usually go wrong”…&… “A bottom line based on average assumptions should be the average bottom line, right? It may miss fluctuations from month to month,” and project to project and task to task.
So given all that why would we possibly want to figure our project thinking on the assumption that that ratio stays constant? You dollar volume is always changing depending upon the task or type of projects you are doing. Why not base your Overhead recovery on the number of Labor Hours your company can generate since that remains more constant and predictable?