Some Thoughts on Overhead

In considering a 'better pricing system,' what role should overhead play? Is it necessary to figure out overhead costs? If we've gone through the trouble of figuring out overhead costs, how should we then apply those costs in a price calculations?

Defining Overhead

I'll define overhead for the purposes of this discussion as, "Business costs not being tracked by some other process." The most common "other processes" that are accounted for independently are materials consumed and "direct" shop labor.

This definition leads us to the answer to the question, "What should we include in overhead?" by saying, "Any and all costs that are not being allocated in some other way." If direct labor is part of our pricing system, then overhead should account for everything that is not ‘direct labor.’ If there is no direct labor element to the pricing system, all costs (besides material) are 'overhead.'

Quantifying Overhead

In calculating overhead costs, one method is to move from area to area and make sure costs in each area are included. Beginning with the office, make sure each person and each office expense is included. Next, make sure that all costs associated with running the shop are included. Finally, look at expenses for the overall facility and the business as a whole. An incomplete list would include accounting costs, advertising, banding/strapping cleaning, computers, equipment leases, insurance, markers, office schtuff, subscriptions, telephone, training, travel, uniforms, utilities, vehicles, and waste collection. Have we forgotten anything? By looking through the bills for the month we might catch some things we haven't accounted for. Our goal in this process is to figure out how much money it costs to keep the business open for any given period of time (a month, a day, and hour.)

The Value

The most important reason to do all this is the awareness we gain of what these expenses are, and the realization that these costs are related to time. If the overhead expenses totaled $50,000 per month and we were only open for business for 5 hours a month, we would only have 5 hours to recover, to "pay for," all those overhead costs. If we are "open," or more precisely "producing," for 200 hours per month, we have to somehow "earn" $250 per hour (beyond labor and material)  we want to pay for that overhead. This is the reason that the most logical way to apply overhead is based on the amount of time the job will take to produce. The argument goes, "Since overhead itself must be recovered based on how many hours the business is producing each month, the only logical way to assign overhead is based on the time it will take the company to produce each job."

We will look at how to do that in later articles.

A Better Pricing System

I've visited a sizable number of component manufacturers over the years and one thing that I've always found remarkable is the wide range of pricing systems employed. To say, "There's no standard pricing model for the industry" is an understatement. Quite a few managers have expressed feelings ranging from ambivalence to outright embarrassment, many saying "It certainly could be better." The goal of this article is describe what "better" might look like.

Right and Wrong

Let's begin with the assumption that there is no right and wrong pricing system, and rather than debate the merits of competing systems, we'll instead describe what we use these systems for. We’ll also describe some desirable characteristics of any pricing system, and then compare those ideas to our own pricing system to see how they measure up.

Features

Here are the basic things I think a good pricing system would deliver:

·         Tell me what price I should give to the customer (naturally!)

·         Tell me the point below which I would start to lose money. I want to know, "How much 'room' do I have?"

Those are the two big ones, but since I'm aiming high, I'd also like it to do a little more for me:

·         Tell me how long it will take me to build. This would help me know how it will affect other work in production and how quickly I can deliver it.

Other characteristics of a good pricing system

A good pricing system is logical, explainable, understandable, repeatable and consistent.

I don't want a system that "works" but no one knows why - it's just mysterious voodoo. Why? If I don't understand why it's working, I won't know what’s wrong or know how to fix it if it starts 'not working.' A consistent, repeatable system would be one that produces the same results with the same input, and is not dependent on a person or ‘impossible to define’ criteria.

Let’s describe a system that’s the opposite, that might work, but not have these characteristics. The statement “This job seems like it has about a 1.7 difficulty factor” might be used in such a system. But if I showed you that same job a week from now, would you still tell me “it’s a 1.7?” On what basis are you arriving at “1.7” as opposed to “1.6?”

Take the Test

So I'll invite you to apply these ideas to your pricing system and see how you make out.

“In considering my pricing system…”

·         Does it tell me what price to give the customer? (Or does it tell me a number that I have to do more operations to in order to arrive at the price I give the customer?)

·         Does it tell me where my ‘break even’ point is?

·         Does it tell me how long it will take my shop to produce?

·         Given enough time, could I explain it to another person?

·         Is it logical; does it "make sense?"

·         Is it consistent and it is used consistently?

If the answer to any of the above is "No," I submit that "it could be better." We’ll talk on ways to go about finding a better system in future articles.

The World Where ‘Eck You A’ Reigns Supreme

Scholars tell us that coins began to be used between 400 and 600 B.C. They were a big hit. Bartering is great, but it has its limits. Coins soon became a widely accepted, universal unit of exchange. Now imagine a world where all component manufacturers have a unit of exchange that is just as valuable as coins are. Because this unit of exchange is completely accepted by everyone, manufacturers can easily compare the production efficiency with each other. Everyone estimates their jobs using the same method. Getting a quote from another manufacturer to build trusses for you is a snap, because you both measure jobs in exactly same way. Oh, and by the way, this is a real place. It is called, ‘Australia.’

Down Under

It’s really true. In Australia, virtually all manufacturers have “bought in” to the concept of converting all forms of work into EquA’s, or “equivalent finks.” Getting deeply into EquA’s can get complicated, but the basic idea is simple. Convert all work into the single unit (our “coin.”) This is done by arbitrarily declaring that “a standard fink truss with one splice takes 1 EquA to build.” The next step is to come up with a method to determine how many EquA’s it takes to set up (as opposed to build) that truss. From there, we have to come up with an easy way to determine the EquA for all truss types. In Australia, this is mostly done through the number of joints, but, as I said, it can get a little complicated. Attached is a PDF describing the method in some detail, if you are interested. My purpose here is not to explain the method, but rather to describe the benefits that the Australians enjoy because they all use the same method to measure work.

The Point

I had heard about this method of estimating before my trip to Australia in 2006. From what I had heard before that trip, the Aussies were a bit ‘EquA obsessed.’ Why would this method be better any other? And how could it really be true that everyone used the same method? During my visit to MiTek Australia I had a chance to talk at length to Andrew Scane, one of the MiTek sales and support guys there. He told me that the small size of Australia (population equal to Michigan + Ohio) had helped. Also, MiTek had promoted the use of this method, and supported in throughout its software suite there.

I asked more questions. I wanted to know if the EquA method was accurate. Was 23 commons really the same as 18.5 cathedrals? At some point, I realized that that wasn’t the point. Once you accept EquA’s as your unit of measure, once you just decide to trust it (like we do dollar bills,) a whole world opens up to you. Once you begin to do everything in EquA’s – estimate labor, calculate production capacity, price jobs, determine overhead, set profit goals, you are now living in a world that is much easier to comprehend and manage.

One Example – The Labor Rate

Once you can measure your work in EquA’s, you can determine plant’s capacity in terms of EquA’s. And once that has been done, you can measure your productivity by combining your overhead and direct labor, and dividing by your plant’s EquA capacity. So if you find you capacity is about 300 EquA’s a day and your overhead and labor is about $4,500 per day, you know your (non-material) cost of production is $15 per EquA. How does that compare to your competition? In Australia, you can just ask, because everyone measures in the same way.

MiTek Australia has written considerable documentation on how to make sure you are accurately calculating overhead, as you can see in the attached document. They actually provide a software program called Profit Centre that allows you to step-by-step enter costs (like those tax-filing programs we sometimes use) and use those figures to really see your overhead per EquA. It really is cool stuff.

Regrettably (I think) in the US, we have the phenomenon known as “no two customers do things the same.” Because of this, most of us have a much more fuzzy sense of our productivity and costs than the Australians do. To be fair, in the US we have a large variety of production equipment – it is more uniform (pedestals) in Australia. Also worth noting, in nearby New Zealand, preachers of the “EquA religion’ have not made many converts. Still, sometimes I wish I lived in a world where we all agreed to use the same “coin.”