Output Thinking: How to Build Systems That Scale a Stage 2 Company

In late October 1991, I was out in the backyard with my son, who was about four years old. We were in Texas, it was 85 degrees, we were sweating, and the leaves were turning brown and falling down. He looked up at me and said, "Is this fall, Daddy?" These were not the falls I remembered growing up in the Northeast. No offense to my Texas friends, there are a lot of pretty places in Texas, but the area where I lived wasn't one of them. So I made the decision to move back.

The problem was that I had a company in Texas and it didn't make sense to move the company. Most of our clients were in the south and southwest. I literally had the thought, if there are companies that have a headquarters in one state and a branch office in another, they can do it, why can't I? This was in the 90s, before the internet made this easy. They used to overnight me data tapes so I could see what was going on at the company.

The company was called Video Rental Services. We sent movie libraries to apartments as a marketing tool. If you rented an apartment, you could borrow the movies overnight for free. We eventually got a website, and you can see how beautiful it looked. That's not actually uncommon for most websites of that era.

The interesting thing was, when I moved, I was able to grow the company and work less. Looking back, I can tell you why that happened. I was able to systemize things and get other people to do the outputs that are required.

You have to be careful when people tell their own stories of their own companies, because it always looks good, especially on Twitter. What I don't always tell you is the time I had to fire my general manager and hire another one from 1500 miles away. I don't talk about the time a competitor stole my marketing materials word for word and my onboarding materials word for word. I don't talk about the time a national company came in, put ads in all the trade journals, that they were going to do exactly what we were doing for half the price. This stuff happens. It's the normal stuff of running a company. But because I could systemize things and get myself out of the day-to-day work, it was the best decision of my life.

The company kept growing. My accountant said, "Hey, move to Russia. It's doing great. Maybe it does better without you." I said, okay. Then I got a call from somebody who said, "You should look into this thing called coaching. You'd be great at it." Coaching hadn't become a buzzword yet, but I got into it and I liked it. I ended up with a bunch of people founding the International Coach Federation. I was their fourth president in 1998. That has allowed me to indulge my curiosity in learning how companies work. I love learning about business models and the different things.

Today I want to share some of the things I've learned from running my companies and from working with clients and learning about their companies. But first, thank you to SMBash for giving you a copy of my book. Some of you were here last year when I was running around with little cards saying, "Give me some ideas about what I should call this book," and some of you helped me with that. If you can take a picture of this QR code, I'd really love it if you go to Amazon and leave a review. We're trying to game the algorithm. It turns out that if you get a certain number of reviews, around a hundred, Amazon will start promoting the book internally.

You are in for an amazing couple of days. You'll wish you were two people because there are two tracks today. There's the searcher track, people looking to buy companies or buy another company, and there's the operator track. How many are searchers? Looks like a little over half. How many operators? About the rest.

I want to say something to the searchers. You have to understand operations because when you buy a company, even if you aren't the operator, it has to operate. Deals are sexy, but the money is made in operations. If it's not running well, it's not going to make your bank happy, it's not going to make your investors happy, and ultimately it won't make you happy.

So I want to focus today on the dilemmas of running a stage two company. Immediately that should bring two questions to mind: what is a stage two company, and why do I care? The reason you should care is that most of you who are buying, new searchers, will be buying stage two companies, and those of you running companies will mostly be running stage two companies.

In order to understand that, we have to take a step back and ask, why would you hire anybody? About 10 years ago, it was popular to say small business owners are the job creators of our economy. I've worked with companies across the United States and Europe. Nobody has ever said to me, "When I get up in the morning and go to work, I want to see how many jobs I can create today." In fact, most of the hassles I hear about, and that I've had in my own companies, come from employees. We would much rather automate. We get contractors. Business owners really don't like to hire employees, even though we like our employees personally. That's not what we do.

Why do we hire people? The job creators are our customers. Our customers demand stuff that they want to pay us for. In order to serve them, we have to get to the point where we hire people. We think, well, I hire somebody to fulfill this role or do this work. But you really don't hire them to do work. You hire them to produce the output of that work. If they could do the output in half the time, you'd be happy.

Outputs are created in two different ways. They're either created by people who are experienced and intuitive and really get stuff done, or they're created by a process, a system. The first one I call "grandma in the kitchen" who has no recipe but makes wonderful food. Holiday meals are terrific, birthday meals are terrific. You cannot scale a restaurant with grandmas. They're hard to find, they're expensive, and if you get two of them in the same kitchen, they probably wouldn't agree on anything. People who build companies like that have a hard time scaling.

The other way those same outputs are produced is by a process or system. "Moving your company from people to process is a maturing stage of a company." I got that line from Melissa Withers, who runs the RevUp Fund. There's a great quote from Daniel Jones, the chairman of the Lean Enterprise Academy: "Brilliant process management is our strategy. We get brilliant results from average people managing brilliant processes. We notice that our competitors often get average or worse results from brilliant people managing broken processes." Average people are much easier to find than brilliant people, but in order for them to produce, you have to have the brilliant processes.

As your company grows, these systems and processes subdivide. You'll have multiple subsystems where the output of one feeds into the input of the next. As these proliferate, you may end up with dozens, hundreds, sometimes even thousands of subsystems in a relatively small company. That's what it takes to move the company forward.

You think, how am I going to keep all this in my head? What people typically do is resort to an org chart. I'd say that's not the best way to do it, because an org chart is people and roles, and roles are a very poor proxy to the outputs you actually hire people to do. If you look at the outputs an average person in your company is producing, there are probably six or seven of them. Some of them only loosely relate to their role or title. But all those outputs are important.

The best way to think of it is that those outputs fall into what I call seven buckets. The first bucket is serving customers. You have to make something or provide a service or product that they want to pay for. There are a whole lot of systems and subsystems that go into producing outputs in the serving bucket. Then you have to find those customers, and you have to sell, convincing them to give you money for what you're making.

Then there are a number of buckets I call the sustaining buckets. The purpose of those buckets is to promote more serving and selling. These are the people bucket, hiring and managing people; the money bucket, paying bills, collecting money, and doing analysis to find insight in your financial numbers; the information bucket, making sure everybody has the right information to do their job; and CYA, a checklist of things that if you don't pay attention to them, they could bite you. It's like getting your blood pressure checked or your teeth cleaned. That's not going to help you run a marathon, but it may keep you from dying when you're doing one. The seventh bucket is scaling, growing the company, being able to sell and serve more.

With that background, we can finally talk about stages. There's a continuum of companies as they grow from stage one, to stage two, to stage three. I got this concept from a guy named John Brown, who wrote a book on exit planning, saying that a stage one company has very little value in the company itself when you go to sell. Stage two has more, and stage three has the most. I've got a piece coming out Monday night in my Substack, you can sign up, it's free, that goes into these stages in much more detail.

How I define these stages is by what the person at the top is mostly doing. In other words, what bucket are they trying to fill? In a stage one company, the buckets they're filling are selling and serving. They are customer-facing activities. They're out there serving customers and finding new customers. They may have employees, but in general, the employees help them do those two things.

At the other extreme, you have a stage three company. Stage three companies are your holdcos, your regional or national powerhouses. The person at the top of those companies is looking to grow and scale. They're looking out into the industry, asking, who should we buy? What market should we go into? What new trends are coming that we need to take advantage of? At both extremes, the person at the top is outward focused. In one case, on customers; in the other, on the industry.

In the middle are the stage two companies. It's not a sharp cutoff. As your company grows, there's not one day you wake up and go, "Oh, I'm stage two now," or "Oh, I'm stage three now." It's not like graduating from college where you've got the diploma. But there are some different things you can see. If you look at which bucket the owner or CEO is trying to fill, that will help you out.

Stage two companies are internally focused, and it's a big shift from being externally focused on customers to internally focused on building the organization. This is a very hard shift. This is why you might want to buy a company at that point, where you can put those systems in place and take it to the next level.

This is also why a lot of companies stay small. I have a 120-year-old house and need a lot of contractors. Most of these people have no website, and it boggles my mind. I get a reference, "Oh, you ought to hire this plumber, this concrete guy." I look them up, they're not there, and yet they get incredible reviews. I say, "Why don't you have a website?" They say, "I don't want any more business. I have all the business I need. I have a crew, I take vacations every year, I'm making some money. I don't want the headaches." They're stage one companies. That's fine, because that shift is hard.

Some of them stay small. Some hire grandmothers in the kitchen who do really good work, no systems, but the stuff is getting done. They become absentee or absentee-ish owners. Many searchers I work with have bought companies like that. The seller is not really involved in the day-to-day, but the people are, and they don't have systems. Now the new buyer has to figure out how to scale this thing. Some business owners just grind it out, working 60, 70, 80 hours a week, and then they get to a point where they say, "There's got to be a better way."

It's okay to stay at stage one. You do not have to grow your company if you don't want to. This works much better if you don't have investors and you don't have bank debt. They may have a different opinion of what you need to do.

The solution to moving from early stage two to later stage two is to build systems and look at the outputs that need to be produced. If you buy a company like this, I assume if you got a bank loan, it's profitable and working, the outputs are being produced. Your challenge is to figure out how they're being produced and systemize them.

This is hard. You can't keep it all in your head. The solution is not to get a big head. The solution is to document your systems. Take that wisdom and experience out of people's heads, out of grandma in the kitchen, and say, "Grandma, tell me how you do this. I want to capture that recipe." There are great tools to do this with. This is particularly hard with knowledge work because it often happens on a computer or in somebody's head, and you have to get it out of them. There are all kinds of tools now with videos: Loom videos, screen share. Those are great for stuff that happens on a computer.

There's a company I know with a cabinet shop, Alex Forbes posts about it on Twitter. They have a thing called Gemba Docs that documents how to run each machine. It puts a little QR code that they print, and the operator scans it and all the stuff comes up on their phone. There are a lot of cool ways to do this, but you have to get that stuff out of people's heads.

The biggest dilemma I see with stage two is that you have to zoom in and zoom out, and zoom in again and zoom out again. You have to go from micro to macro of how these outputs are being produced. You're kind of like a Hoberman sphere, that toy you can expand to about this big and scrunch down to this small, all on hinges. That's what you have to do as the operator of a stage two company.

There's a story motivational speakers tell. You're walking down the street and see a guy with hammer and chisel hitting on rocks. You ask, "What are you doing?" He says, "I'm making bricks." You walk further down the street, there's another guy, looks exactly the same. "What are you doing?" "I'm making a wall." There are always three guys in these stories. The third guy says, "I'm making a cathedral for the glory of God, and it's just going to be amazing." The motivational speakers say you want to be that third guy. You want to do the glorious thing, the incredibly big vision.

I say, if you're building a company, you want to pay attention to those bricks. Because if the bricks aren't square, if they're not uniform, if they're not strong, the wall's going to crumble. The cathedral will never get built. As a stage two owner, you have to go from the vision back to the bricks and back again.

The pattern I've seen is we don't zoom in deep enough. We have this vision and say, "Wow, we're going to build sunflowers from here to wherever." Great vision, very imprecise. How many sunflowers? Millions. That's a big range, but I'd better zoom in and look at what does that one need? What kind of fertilizer, sunlight, water? In my experience, that's not enough. You have to zoom in further, and maybe even further. Eventually you get to where the bricks are made, where the outputs are being produced at that level of detail. Then of course you have to zoom out again and get the vision.

It brings to mind a quote by Herb Kelleher. He started Southwest Airlines. Up until the pandemic, Southwest was the only airline that had a profitable quarter every single year. At one point, Warren Buffett calculated that the entire airline industry had never been profitable. Some companies made a little, some didn't, they merged, some went out of business. Southwest was profitable until the pandemic every single year. His quote: "Think small and act small, and we'll get bigger. Think big and act big, and we'll get smaller." I don't know exactly what he meant by that, but I took it to mean pay attention to the bricks, zoom in, get small, make sure that's happening, and then we can get bigger.

It is these systems that build bricks. The outputs you want are why you have the system. The system itself is how it gets done. There are four parts to every system: the trigger, the input, the transformation, and the output.

A simple example: payroll. The trigger is a day. You usually do payroll every other Thursday or whenever it happens. Other triggers can be an event or a condition, but it triggers the system. The input of payroll is the hours people work, their salaries, deductions. You send that to the payroll company or department, and they do the transformation, taking the input and turning it into the output. They do all the calculations, net pay. The reason you do payroll is for the output: money goes to employees, reports go to management, taxes get filed.

Let's take another one that looks as simple: accounts payable. Maybe you pay bills twice a month. The input is vendor information, who you owe money to, the amount. The transformation is the calculations, cut a check, do a Venmo. The output is money goes to the vendor and your accounts payable goes down. Looks simple, just like payroll.

But then you go, "Wait a minute, I've got somebody doing this. How do I know they're not paying their brother-in-law as a fake vendor?" That's one of the simplest ways businesses get ripped off, a fraudulent vendor. Then you realize, some weeks I don't have the cash to pay all my vendors, I'm going to bump this one out. So there are actually two subsystems here. System one is approving the bills, and the output is "these bills are approved to be paid right now." That's the input to the second system, which actually does the paying. To keep fraud out, you'd like to separate these and have two different people doing them.

What you realize is that there's a piece of judgment between these two systems. As you're working through how people do their work, when you get to a point where they say, "Then I decide," or "Then I make a judgment," or "Then I use my experience," that's a key. That's probably a subsystem. At that point you can break those two things out and do one of two things.

One is, you might be able to train somebody else how to use that judgment. I was talking to a guy in the HVAC business. He told one of his people, "I don't want to order all this stuff anymore. We had 27 vendors, we cut it back to six. I want you to order this stuff." Guy said, "Great, I'll order it for you." He outsourced that system. He forgot to tell him, "You have to keep in mind how much cash we have." So he ended up with like a six-month supply, and $60,000 of money he didn't really want to spend, because he forgot one little critical piece of judgment. That was a simple thing for him to fix: when you do the buying, here's the limit, here's how you apply that judgment.

Sometimes you can't get that judgment off to somebody else. Then what you can do is outsource the first part to the person who has the judgment, and that's all they do. They sit there and make that judgment, but they don't have to do any of the prep work.

There was an orthopedic surgeon in Pittsburgh, passed away a couple years ago, named Freddie Fu. I don't know if he was the first ever to do this, but he was certainly one of the first in Pittsburgh where he had two operating rooms going at the same time. A patient would come in, and a separate crew would prepare the patient and open them up. Freddie would come in, do his magic, leave, and clean up. Meanwhile, in another operating room, another patient was ready. Freddie would go in and do his magic on the second one while somebody else was closing down the first patient. He couldn't replicate his judgment or skill, but he could delegate the rest of the stuff so that's all he did. That's another way to make these systems produce more for your company. Whenever judgment is involved, that's a key.

With the first part, sometimes you can automate it with software. Sometimes you can hire people offshore or in other areas where it's cheaper. So you don't have somebody doing all that other stuff. I was talking to somebody about this and they said, "Oh, you mean the grunt work?" I said, "Well, it's grunt work to you, but to them it's not grunt work." There's a concept that everybody gets to play at the top of their game.

I have a client who's a lawyer. He was hiring paralegals and said, "In my experience, the people who are the best paralegals have come from a tough background, broken out of their upbringing, and achieved a measure of success to be a paralegal. They're great people with that background." I said, "Yes, because that's the top of their game. But if you've been to law school and somebody hires you to do paralegal work, that's not the top of your game, and you feel like it's grunt work." If you put those systems in place, if you break them out, then everybody plays at the top of their game.

Another example from healthcare: another guy in Pittsburgh operates the knee and hip replacement at Magee Women's and Children's Hospital. He has a thing called a patient-centered program where an MD is not allowed to do anything that a physician's assistant is licensed to do. A PA isn't allowed to do anything that an RN is licensed to do. An RN can't do anything an LPN can do. You can see how this is going. He found that patient outcomes were better, and the cost was better, because lower-level people with lower salaries and less training are doing more of the stuff that only they can do, and the people at the top are doing things only they are licensed to do.

If you don't work in healthcare, you probably don't have licensure telling you where these divisions are. But that's your job as the owner or operator of the company, to split that out and say, "Wait a minute, you just do these things, you just do these things," so that everybody's playing at the top of their game. That's where judgment comes in to break out your subsystems properly.

Let me give you another one. Has anybody felt like their email is overwhelming? It's a complaint I hear a lot. "My day is overwhelmed by email." That's kind of like a carpenter saying, "My day is overwhelmed by hammers." Email is a tool to get work done. The reason it looks overwhelming is because it's not really one big thing. It's lots of little problems. The reason I know that is each of these subsystems has a different output.

The exercise I do with people is to say, look through your emails for the last week or two or three and tag them by the kind of output that email requires. You'll probably find it's somewhere around 10. It may be six, it may be 12. It's not going to be 500, even though you may have 500 emails. For each, you want to develop a separate system to produce that particular output. Some emails are sales inquiries, that's one kind of output. Others may be complaints, billing issues, scheduling issues, whatever they are. Figure out the outputs for your inbox, and then you can start taking care of them.

You do that by describing each type of output, making a system for each, and assigning that system to a role in your company. I'd encourage you not to assign it to a person, because if you lock it into a person, it's hard to promote that person, hard to make changes without going back to all your vendors or customers. Put it to the role, so the person can change and the role still produces the same output. Make generic email addresses as often as you can, so somebody sends an email with a billing question to billing@yourcompany, sales@yourcompany, scheduling@yourcompany.

One of the tricks is that the first system of managing your email has to be triage: going through the inbox and saying, "What kind is this? Where does it go?" That's the first system. People have a hesitancy because there's so much info, so much personal stuff in their emails. There's not as much as you think, first of all. Secondly, get a private email. One guy I know has his name twice at his company, and he gives that to family and friends and certain people he knows that's only going to go to him. The other stuff somebody else can do. When they do this, don't ever have them delete it. Have them send the email to another inbox, archive, taken care of, whatever you want to call it, so it's always there and you can go back. Generic email addresses handle it by role. Use software, in Gmail they call them labels and filters, that automatically segregates them.

The last tip: design it for the bell curve. Most things happen in the middle. Don't plan for absolutely every single outlier. Plan for most of them, the 85, 90% of them. But make sure your system doesn't lose the outliers, that when something is an outlier and isn't handled by any of these other things, it goes somewhere that somebody looks at it and can handle it in a timely manner. That's one way you can break your email monster down.

Here are some takeaways. Duplicate systems make your company scalable. When you have systems, you can have more people doing the same thing, and you get bigger throughput. When you document those systems, you get better results from average people, and now those documents become your training program. Other people can do the same thing just as well. When there's judgment involved, that may be a clue there's a subsystem, and you can break those systems apart. Subsystems mean everybody plays at the top of their game, doing the things they are uniquely qualified to do. This also gives you a career path for people, because once you get trained to do the next level, we can move you up. If your company's growing, you're going to need more people at those other levels. Design for the bell curve. The company needs systems. That doesn't mean you have to be the one who creates them. If this isn't how you think, find somebody who thinks this way, either in your organization, a contractor, or bring somebody else in.

Let me tell you the resolution to my video rental company. We sold movie libraries to apartments. As you can imagine, I don't own that company anymore, because how many of you get movies from VHS tapes, which is how we started, or DVDs? This was the second iteration of the website. My wife was running the company by that point and she had better design sense than I did.

The company hit its peak in the early 2000s and was dwindling. I saw the internet coming, I'm not stupid. I said, "This is going to end. We'll just ride this out, take as much cash as we can." It lasted way longer than I thought. By 2016, we'd lost the lease on our warehouse. We'd been there 17 years, still had 130 customers. We were down to two employees. The last year, we basically just made jobs for them. They were the only ones who made any money. But I was reluctant to give up 130 customers. As an entrepreneur, it just goes against my religion. We figured maybe we could get another year out of it, but couldn't find a place that would give us a year's lease.

I said to my son, who lives in Colorado and likes to ski and doesn't like to work as much, but is a great salesman and does a lot of good consulting when he needs to, makes some money, and goes back on the slopes, "If you can find a place," he lives in Leadville by the way, the highest-elevation incorporated town in the country, 10,000 feet, I have to go to Denver for a day to acclimatize before I see him, "if you can find a place that has flexible rent, you can run this company. Get another year out of it. You can do the work these two guys were doing, make the money they were making, and still have time for skiing."

He said, "Great." Moved it up, found a place. He was tougher than I was. There were customers who were a pain, and he said, "I'm tripling your fees, take it or leave it." A surprising number took it. But it kept dwindling. At one point it was down to a group of nine customers, and he had moved it to his house, had a room upstairs. He was working one day a month and making a little bit of money.

In Leadville, rents were going sky high because it's a vacation area with a lot of Airbnbs. The local workforce can't afford to live there for the most part, and he was sensitive to that. He found a local agency who said, "We want to rent your house for six months to serve people like battered women who need shelter for a couple of days or a week. We'll move them in and out, take care of everything, we just need you to be gone." He said, "Great, I'll live in my camper, that'll be fun." That worked for a little bit until his girlfriend said, "Why don't you come move in with me?" That was okay too.

Then he called me up and said, "I have to shut the company down because I have to move out of my house." I said, "You have a business. This is more like a job, a stage one thing at that point, but you work one day a month and you make money. Somebody will pay for that. You could sell this thing." He said, "Really?" I said, "Yeah." He actually sold it. He found somebody to buy the company. That was around 2019 or 2020.

About a year ago, I read that Netflix still had a part of their operation where they were sending DVDs out in the mail, and they were going to shut it down last September. I called up my son and said, "Is Video Rental Services still going?" He said, "Oh yeah, the person who bought it is still running it." I said, "You know what, son? We have outlasted Blockbuster and Netflix DVDs."

So let's stay in touch. Thank you very much. You can find me on Twitter and on email. Thank you for being here.