Boss Life (4 page)

It took three months of shoulder-to-shoulder training before I gave him his own customer, and another month to close his first sale. Once he got comfortable with the idea of selling, he proved to be a steady producer. He was good on the phone with clients. He mastered all the software and learned how to put together an attractive proposal. He realized that there was no trickery involved—that people buy our product only because they want it. And by bringing his shop-floor experience into his own work, he proved to be an imaginative designer, with a lot of good ideas that I hadn't thought of.

By the spring of 2011, Nick and I were selling at about the same rate and I began thinking of hiring another salesman. There's a pleasure in closing a deal and getting the money, but after twenty-five years of sales, I was tired. I wanted to ease myself out of that job in order to concentrate on improving the company in other ways.

But I would not find my next salesman on the shop floor. None of my other employees were good candidates. Out of the blue I got a call from someone who might be a solution to my problem. Dan Smolen had worked in shops in Colorado for years, eventually moving from the bench to management. He was relocating to Philadelphia to be closer to his wife's family. He was calling around, looking for a project manager position. Common in the construction industry, this job involves shepherding a project through production, delivery, and punch-list completion. Dan was used to dealing with clients, although not as a salesman. He has a typical shop-guy demeanor—calm, careful, and competent.

Dan's references checked out and I interviewed him a couple of weeks later. I told him that I needed a salesman who could design projects as well. He told me that it sounded like an interesting challenge and that he'd give it a try.

The offer I made Dan conforms to a pattern I have followed over the years—hire the first person I can find who might possibly be able to do the job. I'm particularly inclined to favor people who show up unbidden and announce that they want to work for me. A number of my best workers, including Nick; my engineer, Andy Stahl; and my shipping manager, Bob Foote, came to me this way. I have found placing want ads to be discouraging. I have to conduct the entire process myself: write the ads and place them, field phone calls, read résumés, do follow-ups, interview, and make the final decision. Technology has eased some of that, particularly writing and placing ads, but the rest can't be automated, and it takes a lot of time.

Furthermore, in the trades, the whole résumé/interview thing is not a very useful predictor of skills. Good craftsmen are generally not big talkers. Interviews don't tell me much about how well a worker might perform. Putting a candidate out on the shop floor for a tryout also presents difficulties. The guys out there are very busy. Who will spend a couple of days with a new person? And what would he work on? I don't want untested workers making anything that customers have paid for. And I can't run the risk of someone getting hurt, or one of our machines being damaged because of an inexperienced operator. My solution was to pick one of the candidates, make the hire, and hope for the best. And, predictably, it did not work out well. Not only was the initial choice a crapshoot, but what happened next was worse.

We don't really have a training program to bring new people up to speed. Steve Maturin, the shop foreman, is in charge of integrating new hires. He's a superb craftsman but not a people person. His method is simple: put the new guy at a bench, give him some work, show him how to do it, just once. If that lesson does not sink in, the guy sucks and is not worth further effort. Steve will do his best to find work for him to do, but he does not run a school. If the newbie is smart, he will observe work being done, ask the other workers, and get the information he needs. If talented, he'll be able to build at a high level. If he is not so talented, or not such a fast learner, he'll flounder until I notice the problem and fire him. As you might imagine, with such a bad process, we have trouble finding new workers.

Before I hired Dan Smolen, all new employees started on the shop floor. I believed that the foundation of success at Paul Downs Cabinetmakers was the ability to build furniture. I considered starting Dan in the shop for a few weeks to give him a sense of how we made our product, but decided against it. I needed him to learn my methods as soon as possible. Whether he was great with a sander or could splice veneer was not relevant. I decided to put him through the same process as Nick: set his desk next to mine and let him see what I did all day.

I had also hired an administrative assistant, Emma Watson, who also needed training. It was time to put everyone together. We replaced a wall between two empty offices with a sliding patio door and moved five desks into that space. Nick and Andy are on one side of the door; Emma, Dan, and I are on the other. If the door is open, we can easily speak to one another. If the door is closed, we can carry on two phone conversations simultaneously. Nobody has any privacy, but we have other offices if we want that. Andy, like me, found that sharing space is energizing and makes communications much easier. He is the link between the sales operation and the shop floor, and being in constant touch with the salespeople facilitated that part of his workload.

Dan started in May 2011. He quickly learned to use our software but had a tougher time when on the phone with clients. It is difficult to think and speak on the fly, to provide instant answers to highly technical questions, even when one is very familiar with the subject. And Dan was still learning about conference tables: what sizes work in which rooms, which configurations work best for which situations, which woods look good together, how to integrate wiring. If he could survive the initial conversation, though, he did OK. His written communications and design ideas were good. He made his first sale in July 2011. By the end of the year, he had closed sixteen deals, worth $270,870. As a comparison, Nick wrote fifty-one orders for $972,601; I did forty-eight, totaling $887,356.

The record for booking orders is held by me, as you might expect. In 2006, I closed 101 deals, worth $1,570,954. I accomplished this along with all my other duties running the company. At the end of 2011, it looked like my record would not be threatened any time soon. Both Nick and Dan could sell, but they were streaky. There would be a week or two when one or the other would be crowing about his successes, and then weeks when nothing came in. I could not tell whether the pattern was something they were doing or whether it was baked into our business model.

We get calls from all kinds of potential clients, and the resulting orders vary in size by a factor of ten, from a couple thousand to hundreds of thousands of dollars. Some callers need fast shipment, and some can wait months or years. So it is not surprising that the timing of orders varies. I had experienced sales streaks myself, but Dan's and Nick's dry patches were more frequent and longer than any of mine. Was it due to some failure on their part? Or was it just chance? And yet, I pay those guys every two weeks whether they sell or not. When I moved Nick from the bench to sales, his biggest fear was that his income would go down. So I offered him a salary that was equivalent to what he had earned as an hourly worker: $62,000 a year. He would also get an additional 2 percent commission on any orders he booked. I gave Dan the same deal. A lot of bosses put their salesmen on 100 percent commission and let them sink or swim. I am not so hard-hearted. Selling is tough. I have been subject to its roller-coaster ride for twenty-five years, and I want to do what I can to shield my people from it. So I wimped out and gave both of them most of their pay as salary, with the opportunity to make another twenty thousand dollars if they got to a million in sales.

—

THE FIRST MONDAY
of 2012 was just like every other Monday. After the meeting, everyone scattered to their work stations. Six on the shop floor, cutting wood and making furniture. One in the finishing room, putting a smooth and gleaming coating on raw wood. Two packing finished work for shipment, with pauses to clean the shop. For the shop guys, each day is much like the others. The projects change, but they are always in the same shop, with the same machines, cutting up the same material, and working alongside the same people.

In the office, it's entirely different. The six of us live the triumphs and tragedies of sales. My mood closely tracks the frequency of inquiries and the arrival of orders. In the first week of January, we get fourteen calls and e-mails, an improvement on 2011's average of twelve per week, but no orders come in. The second week, inquiries remain strong: seventeen strangers reach out to us, each a potential buyer. Still no sales. Now I am starting to worry. Two weeks without a deal isn't unheard of, but as the dry patch lengthens, my discomfort rises. I relive the agonies of days without work, without money. Thankfully, on the morning of Friday the 13th, Nick closes the first deal of the year: an order worth $16,940 to a housing developer in Baltimore. That same afternoon, he chalks up $8,038 for a table inlaid with the purchaser's initials. In the next week, Nick books four more orders, worth $123,986. In the last week of January, he sells three more, worth $25,041. His total for the month is $174,005, a fantastic start to his year.

We will not make our goal of $200,000 in sales, though, because Dan has not sold anything. In fact, he hasn't made a sale since before Christmas. I could also blame myself for not closing any deals, but I hired a second salesman so I could get out of that game. I still pick up the phone when it rings, and if it is a sales call, I talk to the client, do the initial consultation, then pass the job to Dan or Nick. I want to take the lead on certain kinds of jobs, especially when there is the potential for a very large order. I don't trust either Nick or Dan that much, yet. But I send them the bread-and-butter jobs, the ones that might yield from five to twenty thousand dollars.

On the last day of January, a Tuesday, just as I am seriously considering whether hiring Dan has been a mistake, one of his clients pulls the trigger. He closes a deal with a telecommunications firm in Maryland for $18,694. We finish the month with new orders worth $193,602. Not quite my $200,000 goal, but pretty close. And Nick's hot streak continues into the new month, with a $23,327 sale on the third. Our total for new orders as of Monday, February 6, is $216,929. Nick has sold $197,332. Dan has sold $18,694. I took a $903 order for some replacement parts. I'm happy, and I'm unhappy. One of my salesmen is killing, the other is dying. I know that I can step into Dan's shoes on any day, do a better job, and put his salary into my own pocket. Should I pull the plug? It's a tempting thought, but I know better than to act rashly. I've put a lot of effort into training Dan. Better to wait for a while and see what happens.

D
ATE
: M
ONDAY
, F
EBRUARY 6
,
2012

B
ANK BALANCE
:
$125
,
891
.
42

C
ASH RELATIVE TO START OF YEAR
(“N
ET
C
ASH
”): -
$11
,
262
.
90

N
EW
-
CONTRACT VALUE
,
YEAR
-
TO
-
DATE
:
$216
,
929

I deliver a mixed message in the first February meeting. Along with all the new orders, the value of jobs constructed last month was $216,614, and we're poised for strong shipments in February. It's a great performance by the shop. So why do I have $11,262 less than I had at the beginning of the year?

Because money goes out the door as fast as it comes in, often faster. We've received deposit payments for most of the new orders, and preship and final payments for the jobs we delivered in January. That's a nice pile of cash: $215,034. But we spent even more: $226,297. We've had twenty-six working days so far this year, and it's cost $7,803 a day to run my factory. At that burn rate, I have sixteen days of working capital on hand.

I use QuickBooks to make sure that my accounting conforms to standard methods, but its cash management tools are very poor. I rely on the spreadsheet I designed in 2008 to look into the future. Directly after the meeting, I bring it up to date. I add all the incoming and outgoing payments I know about. The sheet is divided into weeks. I can see my cash balance each day, and, of course, each week's ending balance is the next week's starting balance. If I see a negative balance on any day, I'll move payments around. There's always a day in the future when I've used up all the income I know about. The trick is to keep bringing in cash so that this day never arrives.

After completing my update, I can see that working capital will hold steady throughout February, but only if we can book new orders as fast as we did in January. We'll need those deposits to keep up with the current rate of spending.

I don't want to give you the impression that I worry about money all the time. It's probably only several hundred times a day. If I feel like thinking about something else, I make a visit to the shop floor.

It's a big space, 252 feet long, 122 feet across. It has a fourteen-foot ceiling and big windows on the outer walls. Concrete columns, set on a twenty-four-foot grid, break up the expanse. I can see carts loaded with completed table parts waiting to go into the finishing room. Beyond that are stacks of lumber, long rolling racks with shelves of veneer, and then the machines.

There are so many machines that it's hard to spot the workers. The smallest are hand held, the largest as big as a minivan. They fall into three categories: cutting wood parts, sanding them, and clamping them together. When I began my career, woodworking machines were relatively crude and relied on the skill of their operator to prepare them for use. Workers are not consistent about how they do this. Every person who uses a tool ends up with a slightly different result, and it is impossible to duplicate an earlier setup. A project with many steps will start to accumulate errors. A good craftsman makes adjustments as the job progresses, but it takes great skill to produce complex work successfully. The best way to ensure high quality in that environment is to have one person take a job from start to finish, so that he can compensate for these inconsistencies. Another approach is to break the construction into small parts that are built repetitively: mass production. In my shop, we've never sold enough of any one item to need an assembly line, so we stick with the single-maker model.

In recent years I've bought a newer generation of machines. These are called CNC (computer numerical controlled) machines, but they're better called robots. Instead of shop floor setups, they follow instructions in the form of computer code, written by Andy Stahl. We have two machines that operate this way: a laser and a CNC milling machine. The laser cuts very intricate, delicate shapes, like the logos we cut from veneer. The CNC is for big parts. It has a working area five feet wide and ten feet long and can make heavier cuts in bigger pieces of wood. Both these machines can cut to within a thousandth of an inch of the planned dimensions. This means that we don't have to worry so much about accumulated errors. We can design complex work knowing that the pieces will fit together without a lot of adjustments, and we can have multiple guys working on a single project.

Another newer type of machine uses digital sensors to adjust to varying workpieces, but still requires an operator. We have two very expensive German veneer sanders. One can sand the pencil marks off a sheet of paper. A very skilled person with a hand sander can also do this, but not nearly as fast, and with a higher rate of failure.

A surprising truth about machines is that they have varying capabilities, just like human workers. Our second veneer sander is supposed to perform to the same high level as the first, but it cannot. It simply isn't as good: every now and then a part put through it comes out ruined. We bought it first, but now we rarely use it.

There are very strict limits on the size of the material that our machines can process. This has changed the way we design our work. The largest piece that our CNC machine can handle is five feet by ten feet, and our veneer sanders max out at 54-inch width. Many of our raw materials come in sheets that are four feet wide and eight feet long. It's hard for one worker to lift pieces bigger than that, so we use four feet by eight feet as a target for the maximum size of any component, even if the finished table is considerably larger. This is a good limit for other reasons as well: bigger pieces are more expensive to ship and won't fit in elevators when we get to a customer's site. Remember the feedback loops between the various functions of the business? Here's a good example: the needs of Manufacturing and Logistics influence Design and Marketing to limit part sizes.

Most of the time, the salesmen can steer a client to decisions about table size and shape that are within our targets. But would I decline a job if it looks like it will be difficult to build, or ship? Or just suck it up and do it anyway? It depends on the amount of money involved. It's always tempting to close a large deal. But what if the client has a lot of money, wants something tricky to build, and we are not sure whether the budget is sufficient to cover the cost?

There's a job like this on our shop floor now: a big U-shaped table. Each arm is thirteen feet long, and the base is twelve feet wide. We're building it for Company S, a food giant with headquarters in the Midwest. The job took a long time to develop. They first called in the spring of 2011, but didn't give the go-ahead to submit a proposal until November. Nick did the initial design, working from a very simple sketch provided by the architect. The only detail on the drawing, other than the size and shape of the table, was that the top was in three pieces, with a top depth (the dimension from the outer edge to the inside of the U) of five feet. Big problem, as the two pieces forming the arms of the table would be thirteen feet long, and the piece forming the base of the table would be twelve feet long and five feet wide. None of these would fit in our sanders, and they would be hard to move around the shop. Company S also wanted a very traditional look, with lots of intricate detailing on the base of the table.

Nick submitted a first design in November 2011. They thought it was promising but wanted some changes. Nick made some changes. It was better, but they wanted a few more tweaks. Nick made a few more tweaks. Now they liked the design but were nervous about hiring us, since we are far away from their headquarters.

I offered to jump on a plane and make our final presentation in person. In early December 2011, I spent several hours working with the chairman and his team. Face time did the trick. I was able to tease out their biggest concerns: that the table would be late and over budget, and that it would be flimsy. I closed the deal by agreeing that the tabletop would be made in three pieces as originally specified and that the mid-March delivery date was inviolable. And we'd absolutely stick to the original budget of forty-five thousand dollars. The timing was difficult, but not impossible. The budget would be a crapshoot.

We price our tables using a very complex Excel spreadsheet that takes into account the size, shape, materials, number of pieces in the top, power/data configuration, and other factors. It spits out two important numbers: the overall price and time estimates for the number of man-hours needed to complete the job, broken down by design, construction, finishing, and shipping. Our spreadsheet is reasonably accurate for most tables, but for very large and/or very complicated jobs, its algorithms become questionable. When we can't predict how some jobs will go, we guess and hope for the best. We've estimated 314 man-hours to do the Company S table.

Steve Maturin, Ron Dedrick, and Sean Slovinski are building it. All three guys have superb hands. This project requires them to dust off old skills and complete some very difficult tasks without help from our new machines. We're using our CNC as much as possible. The U-shaped top is made of subsections, each small enough to fit on the CNC, that will be glued together to make the three large panels. The entire perimeter of the U is composed of a band four inches wide, made of solid cherry. This outlines a six-inch-wide band of mahogany. The mahogany band surrounds the center panels of the table. These are thirty inches wide and veneered with pommelle sapele, a highly figured African wood. Each of these bands, composed of multiple pieces, will be cut by the CNC machine, then glued together so that the top surfaces are perfectly aligned. Those assembled top pieces will then be hand-sanded to perfect flatness, a task we'd normally do with machines.

If you watched Steve and Ron build this top, which took two weeks, you wouldn't see anything spectacular. No shouts of triumph or end-zone dances. They started with the glue-ups. All those pieces, almost three hundred feet of perimeter, are held together with clumsy iron clamps. Each inch of that length needs to be tapped into perfect alignment, while the clamps are being tightened and before the glue dries. Glue-up takes the first week. Then they do nothing but sand, for hours and hours without pause. They have to move a hand-held sander at the perfect speed or it will burn through the paper-thin veneer and the top will be ruined. They repeat a single physical motion: starting with hips tight to the edge of the table, bending at the waist until their backs are nearly horizontal, arms fully extended, then returning the sander to the start position. Every pass has to be perfect, or the sanding will be uneven and the finish will not look right. Bend. Extend. Straighten. Bend. Extend. Straighten. All day long, holding a hot, noisy sander. This is hard, sweaty work that requires the delicate touch of a concert pianist.

It's a pleasure to watch. Unfortunately, as a businessman, I am not so excited. Highly paid workers using traditional skills to build a complex design is something I never want to see in my shop again. Three of my best workers are bogged down in a single project, and simpler jobs are not getting any attention. We could complete the easier work ahead of their time estimates and meet my monthly build and delivery targets. Instead, the Company S table is sucking up labor hours and leaving my guys too tired to do overtime. We are losing money.

Company S is just one of five orders on the shop floor in February. Their values total $187,639. This is short of my $200,000 build target, but we built $216,614 in January, so I am not too worried. There's other activity out there as well. The tables we built last month are being packed and shipped. And in the office we're processing the usual stream of incoming inquiries, trying to convert them into sales.

Customers contact us in two ways: phone calls and e-mails. The vast majority of these are initiated by someone who visits our Web site. We make a very specialized product that lots of people want to buy but cannot find in local stores. So they head to Google. Within a second, the connection is made: interested shoppers meet Paul Downs Cabinetmakers.

Before the Internet, matching a buyer and a seller was far more complicated and expensive. The more specialized the product, and the greater the distance between client and maker, the harder it was to do a deal. Buying ads in media with national or international reach was extraordinarily expensive. Only a business of substantial size, selling a product that everyone needed, could afford to do this. A tiny company like mine was forced to serve a local market. Direct-mail campaigns and local print advertising were not cheap, but within the realm of possibility. That is exactly how my business developed in its first fourteen years. I spent my marketing budget on print ads in a local magazine.

My first ads appeared in 1993. These fledgling efforts required the services of a professional photographer, writer, and graphic designer, who cost me about ten thousand dollars. (Now it's almost free, using digital photography and cheap software.) Monthly ad charges exceeded two thousand dollars, for an eighth of a page in black and white. I raised the money to start this effort by selling some of my company to my father and brother—a pity investment on their part.

The magazine had a monthly circulation of a hundred thousand. How many of those people were looking for custom dining sets at any moment? Not many. A very small number of potential buyers saw my ad, and a subset of them made a trip to my shop to see the designs I offered. Who has time for that kind of shopping? Wealthy women with large houses to fill.

These face-to-face meetings were nerve-racking experiences. We would go through my portfolio of photographs to see if anything was of interest. There was no guarantee that the client would see anything she liked, nor any guarantee that she could afford it if she did. It is amazing that I sold anything at all. As a feedback loop between Design and Sales, though, these meetings were hard to beat. Clients asked me for all kinds of items: dressers, beds, built-in cabinets, closets, porches, you name it. I made anything, but dining room furniture was the best fit for my capabilities. As a product, it had advantages. First, clients will spend good money on it, in particular to get a dining table made to fit their room. Second, making sets of chairs offers the shop floor an opportunity for repetitive work, which is considerably more efficient. Third, there are lots of ways to improve on traditional designs.