How We Increased Productivity on the Shop Floor

At the beginning of last year, I started using a new technique to analyze my numbers: the trailing sum or average. In a nutshell, you track a set of numbers looking back from the present, at regular intervals, and see how the results change over time. Doing so shows patterns that can be hard to see if you are used to looking at calendar years — it’s as if every month represents the end of the year. It also smooths out variations in shorter time periods. Over all, I find it a useful way to track progress.

Last February, I ran trailing sums on our new contracts and shipments for the previous six months and realized we were outselling our build capacity by 20 percent. Our average new contracts from July to January were running at $213,000 a month, while we were shipping only $177,000. Incoming orders had picked up considerably, but we hadn’t figured out how to match the pace on the shop floor. This new pattern was hard to see looking at yearly statistics, because 2012 had been a disaster, and we had only completed a single month of 2013.

When incoming orders outpace production, two bad things happen: First, our backlog grows to the point that customers become reluctant to place orders. These days, they balk at anything more than a 10-week wait, and they are a lot happier with four to six weeks. So a growing backlog means we will lose sales. But even before that happens, we also lose the potential revenue we could get by shipping as fast as we build. Even though we were making a small profit by shipping $177,000 of work a month, we were leaving another $36,000 a month on the table.

For 27 years, my main worry has been whether we will find new customers. At the beginning of 2013, I was still heavily involved in selling, both in managing my two salesmen and in handling a few clients myself. The numbers showed me that selling was no longer my biggest problem, that I needed to turn my attention to making sure the shop floor could keep up with the sales team. I had two problems to solve. I wanted to find ways to increase our productivity by improving operations, and I also needed to know whether we had enough workers on hand or whether new hires were required.

Walking through the shop each day, I would often see unfinished tables stacked outside the finishing room. It looked as though that step in our building process was a bottleneck. Finishing is a tricky operation, and I have only two workers who can actually do the work at the level required. One of them, Kyle Wegman, had previously been promoted to shop manager. The other, David Dano, isn’t the type of guy to complain about his workload. He would simply stay after everyone had left and keep working.

I didn’t really catch how often he was staying until I looked at the payroll records for 2012 and found that he had worked more regular hours than anyone else (as a relatively new hire, he had fewer paid personal days) and had also racked up 370 hours of overtime — far exceeding everyone else. He had put in more than 2,600 hours on the job in 2012.

His work is physically demanding, not at all like spending long hours at a desk. Clearly, we wouldn’t be able to increase our finishing capacity by making him work more, so I made the decision to hire another finisher. But where to find one? As I said, this is a very particular skill set, and the good ones rarely enter the job market. So we decided to hire and train. This was a gamble, because the job is difficult and requires talent.

My shop manager, Kyle, recommended a suitable candidate for the job, someone he knew well, and whose intelligence, work ethic, and basic skills he would vouch for — his own father, Kevin. That took me by surprise, but I have learned to trust Kyle, and I made the hire. During March and April, David struggled to become an effective teacher, and there were a few rocky weeks while Kevin learned a new skill, but by summer everything was going smoothly.

Just having a second body in the finishing room quickly erased our bottleneck, as each finisher can usually keep up with four or five bench cabinetmakers. Knowing that adding a second finisher would eliminate logjams at the finishing room, I decided to add another person to the shop floor. New contracts were still coming in at a high pace, so I added an apprentice level worker in April.

Along with adding bodies, I also worked with Kyle, the shop manager, to improve operations. We both knew that there were plenty of ways we could get more production from our shop and workers, but how would we identify improvements and implement them? I had come to the conclusion that, as a small company, we simply didn’t have enough bureaucracy to keep control as we revved up our pace.

This is counterintuitive, as rigid procedures and paperwork are generally considered a bad thing for a business — unless you don’t have them, in which case chaos reigns. We found that stupid errors were proliferating all through our production chain, and they weren’t being discovered until we had little time to fix them. Tempers would flare, fingers would be pointed, vows to do better next time would be made, and then it would happen all over again. We never seemed to stick with any of the solutions we came up with.

In the fall of 2012, I had instituted a weekly meeting with Kyle on Friday afternoons. I insisted that this be at the same time every week and that we block out at least 2 hours to discuss every problem that had come up during the week. We both found the meetings enormously valuable. So at the end of that year I decided to add another weekly meeting, this one for the heads of our sales, engineering, production, finishing and shipping departments. We really aren’t large enough to have departments, but I wanted to get representatives of the whole production chain into one room at a scheduled time each week. I called this the Operations Committee.

At the first meeting, I told everyone that our job was to identify problems, experiment with solutions, and make sure that the fixes became part of our standard procedures. The meeting would be held every Thursday at 11:30, no matter what. And I proposed a simple way to track of ideas: Whenever we identified a problem, we would write it down on a 3 x 5 card and thumbtack it to a large piece of cardboard I stuck on our wall.

At the left of the cardboard was a column labeled “Idea.” In the center of the cardboard was a wide area called “Implement.” And at the right side was a column labeled “Sustain.” The plan was that we would move the problems from left to right as we developed solutions, and we would be able to see at a glance the progress of all of the particular items. If nobody had a new problem to discuss at a meeting, we could run down the cards in the “sustain” column and make sure we were still practicing the improvements we had developed.

This was pretty low tech, as you can see in the photo, but it worked really, really well. As of today, we have 23 cards in the “Sustain” column, 12 in the “Implement” area, and eight in the “Idea” queue. Some of the cards note little problems, and some are about big issues: “Do reviews,” “Bonus program,” and “Implement quality tracking program.”

Overall, the combination of the committee and this simple tracking system has brought dramatic change. We have cleaned up the shop floor, ridding ourselves of the accumulation of 12 years of scrap; rethought where we place our machines; and taken walls down so that the shop is now physically much different than it was at the beginning of last year. In the process, we have also changed the way people interact. There is now much more consultation, and arguments have pretty much disappeared.

As the pace of production increased throughout the spring and summer, our biggest problem became administrative. The sales people were too busy to do all of the paperwork their jobs required, and we were having difficulty deciding who should be in charge of the company calendars. So I made one more hire in October, a part-time administrative assistant, Georgia Koulas, who has been such a success that I just made her full-time and put her in charge of all of our scheduling and invoicing.

The result of all of this has been a significant acceleration of production. Our six-month trailing average of monthly shop production grew to $242,736 in December – a 43-percent increase from January 2013.

In the next post, I’ll discuss how my year went and how much money I made.

Paul Downs founded Paul Downs Cabinetmakers in 1986. It is based outside Philadelphia.