Hands on Guide to Cost Modeling
Article 5 in a Series of 5
By Todd Peter
In prior articles the key concepts and structures of developing a Cost Model were discussed. Important ideas to review are:
Granularity: Where in your portfolio of products/customers do you want to be able to calculate cost. The finest granularity is for an individual line item on an individual invoice. Higher granularity can be product groupings, or even at the level of a single plant or office. There is no right answer. The decision rule is your measurement granularity should be the same as your required decision making.
Processes: All products (and services) have a cost related to the business processes used in delivering them. Process cost plus material gives total cost. Determining the processes you want to include is a balancing act between accurate detail and practical data measurement and collection.
Level: This idea refers to how you desire to carve the total cost into different groups. The groups typically are considered incrementally variable, variable, semi variable, semi fixed, and fixed. Selecting the groups properly will give you the best information about cost / volume relationships. The tradeoff is again complexity versus improved insights.
Special Process Cases: we discussed several that can add some twists and turns to the calculation of cost, review article #4 to check if these may impact your situation.
The Universal Cost Modeling FORMULA
At the risk of insulting every reader, the cost formula is simple:
Cost = Sum for each material and process required (qty-each * cost – each). The more difficult part is defining all the materials and processes and defining the numbers qty-each and cost- each. In some cases, qty-each is pretty easy if the engineering department has specified by design the quantity each required and the purchasing department can get vendor quotes clearly defining the cost-each. The amount of data however can be very large with material list running 100’s or 1000’s of parts per product and the number of products in the 1000’s. The Cost-each is usually harder for processes.
The Process Cost, Cost-each Calculation
Here is where the cost modeler earns their keep. Again, the formula is not hard.
Process Cost Each = Total cost for that Process / Total quantity produced.
Your first challenge is to define a process where you can collect from readily available data and data you can get periodic updates the total cost of a process. Your accounting system may not collect cost by process and you must prepare a pre-analysis to collect total cost by process. You may find that your process has a mixed bag of costs at various levels – from incrementally variable to fixed – and this cost must be broken into several pieces to analyze. For example, you have collected the total cost for the “smushing department” for each of the last 3 years. The company ran that department for 1000, 800, and 2200 hours respectively, or in other words the volumes were greatly different. You would need to break out variable and fixed costs to get good measures of total cost for your model.
The second challenge is the denominator – quantity. In some businesses with a single product quantity is a simple sum of the units produced. Once product complexity increases and the required-qty-each begins to vary, then a common basis for quantity must be developed. Often this common metric is hours. But it can be any number of quantifiable numbers such as pounds of finished product, cubic feet of product, square feet of space utilized, pounds of a component product, man-hours, machine hours, cell operation hours, plant operation hours, or some more clever and unique numbers as we discuss below.
The challenge for the cost modeler is finding common units for the denominator that best reflect the nature of the cost/volume relationship. Ideally if the relationship between the quantity in the denominator is 100% linear with incremental costs, pretty linear with variable and semi variable costs, and seems “rational” as a way to split fixed costs. What does the term “rational” mean; an example helps. If you were dividing up the cost of your maintenance expense, using square feet for the process department times quantity produced as the denominator that is more rational than quantity as the denominator. (Now, the perceptive reader should recognize that the term square-ft * quantity is an odd idea. But recall that when we do this for all parts and are dividing the total cost for ALL spreading it by this metric, that while non-tangible, is rational). The process of spreading costs is called in the profession, allocating costs; it is an art not a science.
Choosing good denominators requires skill and experience. The CPA profession prefers the time honored “direct labor hours” as a familiar and defensible GAAP compliant method. And this made more sense in the days of high labor intensity, and a high mix of variable to fixed costs. In today’s world where automation has driven direct labor to relatively very low levels of total cost this can be a very misleading denominator. Just think of a $2M machine run by one operator versus a $100,000 machine by 2 operators. Allocating overheads by labor hours could be very, very, misleading.
Keeping Track of Incremental, Semi-variable, and Fixed Costs is Important
Many business decisions require an understanding of the (up or down) impact on the business from a change in sales volume. We have discussed these costs in earlier article #3. To improve the quality of your cost model you should group and maintain all the costs in these buckets.
So practically what does that mean? A single cost element = qty-required * cost-each. So, as you develop the cost-each for incremental costs it should include all the costs that are tied to the element. Take labor as a familiar cost element. You pay your workers $16/hour. You also pay their payroll taxes (10%), some health care costs (8000/year). Overtime is not allowed (for simplicity). Their labor cost each hour is $16 + $1.6 + $4. While health care seems a “fixed cost” and many firms include it in SGA overhead unrelated to direct labor rates, the discerning analyst will understand that changes in employee counts are going to drive health care costs. This simple example must be repeated for every process in your cost model. Maybe you have one labor rate? Maybe you have some employees at $16 and some at $28.
A well-founded knowledge of incremental, variable and fixed costs components of a “product” is critical to understanding margins and impact the quality of all 10 of the reasons costing is important (article #2).
Costing is hard; it has taken 30,000 words to just scratch the surface! Unlike financial accounting, there are no GAAP rules that dictate right and wrong ways to cost. The goal of the cost modeler is to provide costs and margin estimates, that are close enough to allow prudent tactical and strategic decision making. In many small businesses, owners “know” deep in their bones what makes them money and what does not. At some point in the maturation and growth of a business and especially should an owner desire to sell the business to third party, costing and margins become critical.
I would argue that margin % is likely the single most important guide to the health and future success of a business, and understanding cost and how they impact margins a critical skill for business success and survival. Thanks for reading. Call us if you need any help.
Todd Peter is a FocusCFO Principal based in Cleveland.