More Thoughts on the Cost of Problems
Our last post about the cost of problems drew a lot of discussion on LinkedIn groups. Predictably, some comments related to problems as a cost of doing business and not worth measuring. One person felt that there is too much focus on problems and not on what companies are doing right. While we agree that externally (with customers, shareholders, regulators, etc.) you want to emphasize the positive, internally you need to understand what problems cost so you can put priority on the most expensive issues.
There were also some great suggestions about measuring the cost of problems and taking ownership. One person pointed out that many large manufacturers have implemented smaller work groups of 150 or fewer employees with their own profit-and-loss responsibility in order to focus greater attention on ownership of problems.
In the LinkedIn posts we asked how readers measure the cost of problems and the answers fell into two primary buckets:
- The value of production lost while a problem is being fixed. You determine how much product can be produced per hour under normal operation and the price the company would charge for that amount of product. Then you multiply that by the number of hours the line was down while the problem was being identified and fixed. This takes into account all the costs that would have been associated with production. To get to the true cause of the problem, you have to deduct the value of raw material that was not used and the profit that is factored into the sale price of the product, and you need to add the cost of the “fix” for the problem.
- Actual out-of-pocket cost. This method requires adding up the actual costs incurred while the problem was occurring and/or the line was down, including scrap, lost work hours, cost of the fix, allocated equipment costs, etc. As many people pointed out, this involves getting the accountants involved and not something most want to deal with.
While neither method is easy, there seems to be a preference towards the value of product not produced while the problem was worked on. But neither approach captures the damage to customer relationships because of poor quality or missed delivery dates. Also, most of the time the real cost of a problem is only known after it is fixed and it is known how long it took to diagnose the product and make the fix, the cost of material, etc.
Here is one example from our consulting work about how a client measured the cost of a problem and the results it achieved by finding and fixing the root cause of the problem. The client saw a decrease in output because of a problem or problems in the first steps of the production process that was shutting down the line every few days. After narrowing down the cause of the problem to two possible causes, the team identified the most likely cause and installed new parts costing $300. After implementing the fix, production exceeded expectations at a time when customer demand outstripped production capacity. The company estimated the problem had reduced output by 3.5 million units per year and the value of those units was $350,000.
Measurement of costs like this get everyone’s attention.
Sometimes a problem is so big it needs immediate attention. A line is down and the problem has to be fixed. But there are often problems in a plant that are not life-or-death but are costing real money. The company must attempt to quantify the cost of these problems so it can focus on the problems that are costing the most money and set priority on which problems get attention first.
Imagine if a company had made rough estimates that three problems in a plant were costing $15,000, $100,000 and $350,000, respectively. Which one do you think it would focus on first? You bet it’s the $350,000 problem. If you believe in the 20/80 Rule, then 20% of problems are causing 80% of problem costs. Doesn't it make sense to know which problems that are in that 20% (even it's a rough estimate) so you can focus on them and not waste time fire fighting problems that have little financial impact?