The biggest constraint we Theory of Constraints consultants find is … (part 2)


The Allocation Mind-set

By Dr Lisa Lang

We have worked with companies around the globe and the constraint is always the same.  It’s how we think.  In particular, it’s how the business owner or leader of the company thinks. Last time, in Part 1, we discussed the efficiency mind-set and how focusing on efficiency can lead you astray.  I made the case that efficiency is NOT a precursor to improved performance, but a by-product.  In this installment I want to discuss another type of wrong thinking – the allocation mind-set.

You buy the same equipment as your competitors. You hire from the same labor pool. The only difference is how you think. Unfortunately, you and your competitors also think the same way. So you are left to compete in a market where, from your customers’ perspective, you’re all the same. So they make decisions mostly based on price.

Let me explain the NEXT common way our thinking goes wrong and the negative effect this wrong thinking can have on your business.

The allocation mind-set is where we believe that in order to ensure we are going to make a profit, we have to allocate some portion of our overhead to “product cost”. The idea is that if every product we sell absorbs some of our costs, then we will know at what point we are making money and we can better ensure that we cover all our costs.

So when we calculate the “gross margin” (GM) of a product it looks something like this:
Selling price:       $100
-COGS:                -$60
Gross Margin:    $40 (also called Gross Profit)

Where COGS (Cost of Goods Sold) typically include raw materials and the direct labor used to create the product or deliver the service.  (Some companies may allocate more than direct labor, but this is the most common allocated cost.)

But if you think about it, direct labor really is NOT a variable cost, unless you pay piece rate. And this is true for both manufacturers and service providers (again unless you pay piece rate – which is very rare). You are going to pay your employees this week whether you sell something or not.

It is this allocation of direct labor to COGS that is what I’m referring to as the “allocation mind-set”. The amount of direct labor allocated to a product/service is usually based on annual volume assumptions and the estimated time a particular job will take.

This means that the allocation of direct labor costs to a job or opportunity influences your decisions:

·        Which jobs/projects you take.

·        Which markets you go after.

·        Which customers get preferential treatment.

·        How much you charge.

So far, you’re probably thinking – yeah, that’s what we do, what’s the problem? The problem is that the allocations you do are based on a volume assumption and time estimates. Both of which we know one thing for sure about – they are wrong. The question is by how much and in which direction. 

Not only will the amount you allocate be wrong, more importantly, it can lead you astray. The best way for me to demonstrate that is with an example.

Let’s say that you have a customer who wants to give you more business.  They are one of your best customers and in exchange for the additional business they want a volume discount.  The volume discount is reasonable and something you do all the time.  The problem is that the way they want the product delivered along with their low inventory requirements it’s going to require you to do 3X as many set ups as you would normally do for that volume. 

Using the allocation mind-set you would calculate the gross margin of this new business. And you would allocate the additional setup time to opportunity. Your COGS would include the cost of the additional setups.

Now, let’s say that the result is that the gross margin percent is slightly NEGATIVE with these additional setups. What would you do? Pass on the additional business? Take the business but give that customer lower priority and complain about that customer every time you run their job? What decision will you make with this cost allocation mind-set?

Who knows since this isn’t a real situation, but before you continue reading, please give it some thought. How do you generally feel about more setups or about lower margin work?

If you’re like most people you would probably pass on the business or try to negotiate with your customer to take more products at once so that you could reduce the number of setups you would do. And you may even find yourself saying “the cost of those setups makes this business unattractive for us”.

Let’s challenge our thinking with Theory of Constraints and Throughput Accounting concepts.  Let’s challenge the allocation mind-set.  First, how much throughput would the additional volume generate?  Throughput = Sales – Truly Variable Costs.  Truly Variable Costs (TVCs) are all the costs you pay as a result of selling one more.  Typical TVCs include raw materials, purchase parts, outside services, subcontracted services, freight, and sales commission.  The Theory of Constraints definition of TVCs do NOT include direct labor unless you pay piece rate.  So when we calculate the “throughput” (T) of a product it looks something like this:

Selling price:      $100
-TVCs:                  -$20
Throughput:       $80

Next, determine if you will need to increase your fixed costs (operating expenses) if you take this additional volume? Will you need to hire anyone or buy any equipment? If so, how much?

Let’s say we do not need to hire anyone or buy any equipment. And if this is the case, we don’t currently have an internal constraint.   We most likely have a market constraint.

The way we recommend you think about this decision is by comparing the change in Throughput (ΔT) versus the change in Operating Expense (ΔOE) as a result of this additional business. And if the ΔT is greater than the ΔOE, the difference goes to covering all your operating expenses and helping you make a profit.

The fact is that, in most cases, a setup doesn’t cost ANYTHING (or they cost a little raw material to get the machine lined out).  They do, however, take time.  But it is imperative that you differentiate between cost and time.  If additional setups would consume so much capacity that you would need to add equipment or people, then it would be reflected in the change in OE.
But to deliver that offer you need to do more setups.  But because YOU understand that set-ups do not cost anything you are willing to do it.  And your competitors are not!  Which means you can make thinking differently pay off by taking market share.

I’m not saying that if ΔT > ΔOE that you must take the business, but I am trying to get you to look at the real situation and understand the real bottom-line effect. Because in this example we would potentially pass on business that would add incremental Throughput and if you do have a market constraint then don’t you need more business? Shouldn’t you be trying to determine what you would need to offer your market to take market share instead of trying to talk your customer out of doing more setups?

The allocation mind-set has you striving to reduce YOUR setups so that you can reduce YOUR costs. Notice that it’s all about you. Not a good place to be if you have a market constraint.

Now imagine that it’s not the customer coming to ask you to increase your set-ups, but instead you created a Mafia Offer that better served your customers needs by:

·        increasing availability of the right products

·        by reducing overall inventory

·        by reducing the amount of cash they have tied up in inventory

You can hear an example of such a mafia offer here:

So if you improve your operations by eliminating the efficiency mind-set we talked about in Part 1 and then challenge your thinking about allocations to create a great Mafia Offer – what would happen to your business? How much more money could you make?

The combination of our Velocity Scheduling System Coaching Program with our Mafia Offer Boot Camp will accomplish just that.

In Part 3 we will cover the cost mind-set. If you have questions or comments on Part 2, please click on the “add one” link in the comment section below.

Wishing you Success!

Dr Lisa

(c)Copyright Science of Business, Inc. All rights reserved.

Share and Enjoy:
  • Twitter
  • Facebook
  • LinkedIn
  • email
  • RSS

{ 6 comments… read them below or add one }

Bill Lukens

I’m reading part Two and you say that under the Theory of Constraints Truly Variable Costs don’t include direct labor. How so? You state I’m paying my employees whether I sell something or not. My employees don’t sell, they make and we don’t pay them to make nothing. If we’re required to make more, direct labor is a variable cost. Please tell how it is not.

You state that we base allocations on assumptions and time estimates…that’s us, but I don’t believe either are wrong as you say. We’ve been in this business 52 years and we have a pretty good idea what it takes to provide the services we offer our customers.

I’m trying to understand, but I’m having difficulty following your example. We have jobs, one or two large parts, that take as much time to set up as they do run. We also have jobs that can be set up is several minutes and then run hundreds of parts. Normally I’m more concerned about run time than set up, hence the belief that direct labor is a Variable Cost.

Explanations please.



Hi Bill,

Re: Truly Variable Costs

The definition of Truly Variable Costs are those costs that you pay as a result of selling one more part. And unless you pay your people piece rate they don’t vary by the one. You are going to pay them for the hours they work regardless of how many parts they make. Therefore, by definition, they are NOT Truly Variable Costs (TVC), but what we call Operating Expenses (OE).

Operating Expenses are your fixed costs. They tend to be period costs. When they change they tend to increase or decrease in a step wise fashion. If you triple sales this year your OE may need to go to a new higher level. If your sales drop by 50% (knock on wood) you may need to decrease your OE to stay profitable. The fastest way (in the US) to decrease operating expenses is to cut people.

When you work overtime, your OE increases in that period. And if your employees make nothing or all bad parts you are still going to pay them (you may eventually fire them, but you do pay them). So that’s why direct labor is categorized as an OE in Theory of Constraints.

Re: Allocation

To allocate direct labor to a part you have to know your annual volume, annual labor costs, and the time estimate for the part you’re going to allocate to (all ahead of time). All of those estimates or forecasts are likely to be off to some extent. And, more importantly the exercise is a waste of time. Historically we have allocated to ensure we cover all our costs and make money. But allocation is NOT necessary to do this and in my opinion is old technology. You don’t need to allocate to ensure you make money.

Re: Running parts vs Setup

The key is to differentiate between costs and time. Setups take time but unless you ruin some material, they don’t cost anything.

The Throughput you generate on a job is the same no matter how long it takes you to do the job. Or whether that is more or less than what you estimated. It is our job as managers and supervisors to make sure that our Throughput grows at a faster rate than our Operating Expense for the company as a whole. And if that happens we are becoming more productive(and more profitable) – we are getting more done with the same or fewer resources (people/equipment). And as soon as that happens or we sell more or less then we forecasted or the job takes more or less time – our allocations are all wrong. I’d rather spend my time increasing throughput then getting more precise allocations.

It’s really a mind-set issue if you believe your direct labor costs are variable, they tend to be. But what if you could do more with the same people?

I hope that explanation helps. If not, maybe I should do another video …


Bill Lukens


Thanks for the reply. I’m still having trouble buying the concept. You state the Truly Variable Cost is the cost we pay as a result of selling (making) one more part. That’s all we do. We are basically a contract machining company. We do not make a product. We are a service industry making contract parts; big onesies – twosies (tons) to small parts you can hold a dozen in your hand by the hundreds. If our backlog goes up, we work more hours, if it goes down we work fewer hours. If it goes down far enough, we layoff. If opportunities overwhelm our capabilities, we have to farm the work out, or turn it down. It varies weekly depending on sales and schedules…’s variable. Unlike you suggest, our direct labor people don’t come to work with nothing to do. You are correct however that they get paid for making bad parts, but that only goes so far and then they get to make bad parts for someone else.

If I told my people, “set up is free, you come in and do it for nothing, because it’s free”, I suspect I would be told where to go. If I said, come on in I’ll pay you to do whatever how do I bill my customer for that time. We sell time, parts are simply the unit in which we bill. Time is money. We don’t pay for parts, we pay for time. If we get a time and material job, that’s how we bill it.

Efficiencies: Because we sell time and almost everything is different from day to day, sometimes hour to hour, it is very difficult for one man to run more than one machine….ala you hire a person to drive a car crosscountry. You want two cars going two different directions, it takes two people. Three cars, three people. Very rarely are all cars going to the same destination, if they are, then we can put them on a trailer and only hire one driver. Having said that, we’re always looking fro more efficient ways to set up, machine, inspect, even the processes in the office to set jobs up.

Our machines carry different overhead rates. Our big mills costing nearly a million dollars to install demand a rate different than an assembly person. I don’t know any other way to cost/estimate a job other than to calculate the material and contracted services plus the labor hours carrying different labor and overhead rates. Many times, our customers want labor and material broken down.

Allocation: We don’t need to know annual volumes to allocate direct labor, we allocate it by the job. We bid everything we do. It’s all different, custom if you will. There is no other way to price it other than time and material. If it’s a repeat, you can obviously see how you did the last time and bid accordingly.

I have read the goal and taken the course. It is easier to find constraints when you make the same product day after day. You make the same meal every day, you get pretty good at it. You make something different everyday with different ingredients, it’s more difficult to manage, especially if you need a pinch and must buy a pound that can only be purchased in a speciality store.

Throughput …back to the kitchen.. pop soup in the micro wave, not so a souffle…time consuming. A five course dinner costs more to prepare than a Big Mac….Time and material. Someone has to figure how to price the menu….unless it’s fixed with the same menu everyday, I’ll bet it’s got variable labor in it.

I guess you’re right, I have a mind set! I want to be open minded, but ….





Your shop sounds like many that I work with.

When you sell one part …

You now need to make one part.

You paid or will pay for the raw materials and outside processing (if any) for that part.

If you had not sold that part you would not have had to purchase the raw material and would not have incurred the outside processing costs. Those costs, the Truly Variable Costs, are only realized IF you sold and make the part. (Agreed?)

This is not true for labor. You will pay your guys this week whether they do that part or not because they get paid by the hour. Your labor does not vary by the part, but by the total amount of work you have. (Agreed?)

Your labor can and does vary based on the total work load you have. And because it varies in this way, we categorize it as Operating Expense.

I can’t quite tell where I’m not being clear, so let me know where we agree and where not and why not.

Dr Lisa


Jeff 'SKI' Kinsey

Lisa, most of us know what we know.

When I was a cost accountant in a manufacturing environment, it made perfect sense to allocate overhead (burden some call it) in order to arrive at a cost, in order to calculate a selling price.

In my experience, until each person works through the P&Q Example for themselves, get the wrong answer (which as you know, EVERYONE does!), and reframes the exercise enough times to actually solve, I cannot help them. Your mileage may vary. {grin}

Changing one’s vantage point requires trust. Trust develops into habits. It is extremely difficult for humans to break old habits. Without trust, new habits cannot form. Because the IRS endorses GAAP (and soon, if not already IFRS [I no longer count beans and do not stay current in such matters]), business owners trust it and often view operations readily through this “lens” so to speak. Too few people understand that most cost accounting activities are facing rearward at past performance.

Throughput Accounting (where TVC, totally variable costs, makes sense) looks to the future and is heavily rooted in what we commonly call “managerial accounting” which as should be obvious from the title, is designed for managers that must direct (a forward looking vantage point for sure!) and control one or more processes.

Therefore, for everyone struggling with Part 2 of this great discussion, please consider checking out Goldratt’s P&Q Example. A quick google (using quote marks for best results: “P&Q Example”) will render at least two great explanations.


P.S. Thanks again for another thorough explanation of a hot spot where the wheels come off too large a number of implementations.



Hello Lisa,
I have to agree when you use the mindset that true variable costs are just linked to the sale of the product. A perfect example is the risk inherited when making the product and it has failed the specifications required to deliver. Multiple examples of this are when a part fails at final inspection; a defect is revealed and the part is rendered as scrap. In this case, direct manufacturing and material costs have been separated from the cost of sales and the part made is then lost and there is no sale!


Leave a Comment


Previous post:

Next post: