A Lesson on Balancing Scaling with Stability

Welcome to HBR on strategy, case studies and conversations with the world's top business and management experts hand selected to help you unlock new ways of doing business. In 2013, Chewy.com was growing fast. But its third party logistics partner couldn't keep up. CEO Ryan Cohen faced a dilemma. Slow growth in order to maintain the key relationship or go all in on building Chewy's own fulfillment network. Cohen chose the latter riskier move and it paid off. Chewy.com became one of the most successful e commerce Companies in the U.S. harvard Business School professor Jeffrey Rayport wrote a case study about Cohen's pivotal decision. He talked about it with host Brian Kenny in 2022 in an episode of Cold Call.

So why don't we just get started? I'm going to ask you to tell us what the central theme of the case is and what your cold call is to start the case in class.

Well, the case presents a very interesting conundrum faced by the founder of the company. Many people will know him because he's considered the grandfather of meme stocks these days, but none other than Ryan Cohen, who was one of the co founders of chewy. In late 2013, Ryan is running the company as CEO. It's a 22 and a half year old venture. He is seeing astronomic top line growth and his distribution partner, it's called a third party logistics provider or A3PL is not owned or controlled by Chuy. It's a non contractual handshake based relationship and he has a problem. And the problem is that the Singular 3PL in Mechanicsburg, Pennsylvania cannot keep up with Chewy's pace of growth. And the question Ryan faces, which many, many entrepreneurs who deal certainly with physical goods face, is at what point do you insource big aspects of operations like logistics and fulfillment and when does it make sense to outsource them? Maybe even in perpetuity, depending on the dynamics of the business.

Uh huh, uh huh. And so how would you start the class? What's your cold call?

Well, it's an interesting one because it's a bit complex. I start by throwing in every single person in the classroom into the deep end. The deep end being not just the question of sort of what do you do in a general sense, but quite specifically what do you do? And Ryan with his board of directors with whom he's going to meet, of course it's an HBS case. So there's always a board meeting around the corner. He's got to sit down with his board of directors who are telling him that if his 3 PL can't keep pace with growth, he's got to slow growth down and improve relations with the 3PL. Basically treat it as a partnership and help them come along, because right now they're in a sense, falling down on the job. So Ryan has four choices, one of which is what the board wants him to do, which is slow down and ameliorate that situation. And then he's got three choices related to maintaining the pace of growth, one of which is to go to another part of the country, say the west coast, and establish a second three PL relationship. The second is to start, if you will, the journey down the learning curve to figure out how to run your own fulfillment logistics in house, meaning to go build a fulfillment center on the west coast or in some other part of the country, or the big bet, which is jump in with both feet, sever relations with existing 3 PL, and essentially do a flash cut and insource all of fulfillment. Essentially.

Now, yeah.

And I say that this is complex because it's not just that there's door numbers 1, 2, 3, and 4, but every one of these has a mountain of negative arguments against it.

We're going to talk a little bit more about each of those as we get further on in the conversation, because they're all very complex and, you know, there's good and bad to each of them. I'm curious as to why you decided to write this case. You're not a current pet owner. What intrigued you about the Chewy situation enough to write a case about it?

Well, it's a funny story. I was moderating or helping moderate a tech conference down in New York. Actually, I can say it was a very cool tech conference. It took place in Brooklyn, not in Manhattan. And we had, at the end of the day, a panel which relates to something that's on a lot of people's minds, minds. The question of, in the age of Amazon, are there opportunities for others in this thing called online commerce? And so we had a predictable set of extraordinary individuals. The founders of Warby Parker, one of the founders of Wayfair. We had Ryan, who was founder of Chewy, and one other. And I was asking the inevitable question, which is, you know, after hearing from them about their business models and why they thought they were defensible, I said, tell me what you're really doing to differentiate yourself from Amazon. And Ryan, who is sitting to my left, put his hand up in a sort of polite and slightly sheepish fashion and said, oh, I can give you one. I said, okay, share it, please. And he said, last year we wrote 5 million handwritten notes to our customers. Now that got my attention. It got the attention of everyone else in the room as well as the other panelists. And so after that I called them up or we had a little chit chat and said, that is extraordinary. I've got to understand how it is that you've taken this idea of high touch service to and scaled it on a technology platform in a way that has created again competitiveness, defensibility for Chewy.

For those who are listening who may not be familiar with Chewy, I think they get the sense now that this is an online pet supply store. But I'm going to ask the question in a slightly different way, which is what business would Chewy say they're in?

It's a fascinating question because one version of it, this is not what Chewy would say, but just among us here in the HBS community, I mean, have we ever seen a more commoditized category of essentially grocery like goods? And how do we know that? Well, the single largest channel in the country for pet food and supplies is the grocery channel. Amazon is clearly a huge player. And so the question is, how do you make money selling commodity goods online? You don't, unless there's an angle of differentiation. And hence Chewy would say that they are in the business of taking that wonderful, very personal, often even emotionally compelling experience that pet owners have with their local pet store. I mean, there are 18,000 pet stores in the country. 59% of them are indie stores as opposed to individual outlets of large chains. And in those stores, the owners get to know the pet owners and they recommend goods and they hand sell things. Chewy's idea is, what if you took that and you could scale it in a way that felt so personal and so compelling, even though they're running a platform that serves tens of millions of us consumers.

I've teased a little bit in the opening about how much we spoil our pets. I'm guilty, I think, of spoiling my pet, but not necessarily to the extent that the case describes. Let's talk a little bit about human grade dog food. The thought of that is a little scary to me.

It is. I always find it funny that when I read the description of the ingredients of that private label that Chewy has created called Ty Lee's brand, Ty Lee being the name of Ryan Cohen's teacup poodle, it sounds, I mean, if you can get past the beef heart, which is one of the early ingredients, it sounds absolutely delicious. It's organic, it's grass fed, it's absolutely human grade, no question.

So I would imagine that if you look back 10, 15 years, people weren't buying those kinds of products for their pets. So we've seen a change in the way that we think about our relationships with our pets and a willingness to spend a lot more money in a way to humanize them, I guess, and to really have them be part of the family in a different way. What does the market look like and how much are we spending on these kind of things for our pets?

It's a huge market. As you mentioned, it's over $100 billion today at the time of the case. And we wrote this case quite recently, but went back in time, as you said, to 2013 to frame a sort of existential moment of truth for the company. But back then it was a $53 billion market. The average pet owning household spent about $522 a year. Interestingly, to your point, Brian, that was a big change from where we'd been 10 or 12 years earlier. One of the big questions that we get into is what's the difference between Chewy's prospects and those of pets.com, which was the poster child for.com excess in the first Internet boom? One of the big changes that took place over those 10 or 12 years was what folks in the industry refer to as the humanization of. It's what you talked about, of the furry friend becoming a member of the family. I grew up in Ohio, didn't know about the Barclay Hotel. Now that I know about it, $95 a night, it sounds like a great deal. Maybe I'll try staying there. I mean, what a remarkable place. But over that period of time, the needle really did move. And what I mean by that is that folks who study this market from the standpoint of consumer psychology define consumers in three categories. They are the humanizers of pets. And it sounds like you are proudly one of them.

Absolutely.

They're the folks who sit in Iowa, go to 4H fairs every summer and think of their pets as livestock. And obviously those are the non humanizers. And then you have this wonderful category, about 5% of the extreme humanizers. And these are the folks who not only buy the food and buy the medicated food and buy human grade, but in their effort to humanize the pet, when it comes time for Halloween, the pet needs a Halloween costume. And by the way, Christmas sounds good. Let's get a costume for that too. With more consumers humanizing their Pets. A second thing happened which is referred to as the premiumization of of the product and that leads you to human grade. And I did want to share with you one of my favorite statistics from the case and I will read from this Brian, if you'll forgive me, please. I love the fact that you can go on Chewy and buy Nestle Purina benefit real beef Dog Dry Food, £15 $11.39 Price per pound. My little calculation. This is high level statistics and analysis. Coming you from Harvard, not MIT. $0.73 a pound. You can also buy K9 Naturals Beef Feast, raw grain free freeze dried dog food in an eight pound package, $195.99 and according to my math, call it arithmetic, that's about 24 and a half dollars a pound. Now I don't know about you, but I can't find a lot of meat I can buy at whole foods for 25 bucks a pound.

So ribeye steak I think is about the same.

Right. You can't spend money like that on human food. So the point is we have done something very significant to change over this 10 years, both the customer and the product and hence the economic profile of the business. Meaning that with higher prices there are more gross margin points, there's more mar to play with and all of a sudden maybe you can build a business that's worth something.

We haven't really talked about the big box competitors that Chewy deals with. These are places where you can walk in and all the products are right in front of you. What does the competitive landscape look like when you bring those guys into the mix?

Well, they essentially split the market with the Grocery Channel and Amazon. As I said, there are many, many of these indie stores. But alas, the impact of the big boxes in online commerce has meant that the independents who represent 60% of stores for round numbers, represent only 4% of revenues in the industry. So the industries owned by the giants, Chewy was in effect going up against not just Amazon and the grocers, but as you say, these very well established chains that each had between 1100 and 1200 stores in the US at the time and billions of dollars of top line revenues.

Yeah. Now, Pet supplies or pet food was not the first idea that these partners had when they started off with a business. They were going to do jewelry, which is obviously quite different. Like how did they start there and end up with pet supplies?

You know, there's that famous story about Jeff Bezos driving across the country with his then wife Mackenzie Bezos going through Analytically, two dozen categories of e commerce and landing on books as the right thing to sell. If you are an e commerce pioneer. In a sense. Ryan and his two co founders, Blake Day and Alan Atal, did something similar. They asked the question, which is maybe an obvious one for anyone who's been in the online business for a while, which is what is easy to ship high in value, hence rich and gross margin points for which the logistics are obviously more manageable than they would be if you were Wayfair shipping sofas.

Got it.

So they came at this initial selection of the jewelry business very analytically until they realized two things. One is that the existing players who weren't big enough to clobber them, but on the other hand were big enough to demonstrate that there was only so much scalability in that business and that the economics were not so attractive for the simple reason that there was very little in the way of repeat purchasing. They then made the famous pivot.

Let's talk a little bit about the early days of chewy.com and the sort of situation that they faced financially. This was not a well financed operation initially. Can you talk a little bit about the way that they were able to sort of bootstrap this together and get started, at least on the path that they're on now.

They were running the business essentially on what's referred to as a purchasing card. It's a kind of credit card for which you don't revolve credit. You've got to settle it every week and it's granted only to businesses. And they had an $800,000 credit line that they were maxing out every single week. So this business was in effect running on fumes. Just before the time of the case, they'd managed to raise $15 million from in fact, a Boston based VC called Volition. Volition was a believer. A few private investors came in, but Chewy was burning half a million dollars a month. It had $7 million of cash on hand. So they had 14 months worth of Runway. And this is back to where we started. This is an awful position for any scaling startup to be in. To have limited Runway, limited access to additional cash, to have essentially a business where you're taking title to the inventory and trying to sell it as fast as you can, either before or after you take the order. Very challenging to make a business like this work.

But at the time of the case, they were actually pretty well established at this point. And so they faced these four decisions that you described. Can you talk a little bit about the challenges that they were facing with the three PL that they were working with at the time.

So this was a 3 PL that knew something about E commerce but knew nothing about shipping big and bulky. And the result of that was that packages were wet, they were packed in a haphazard fashion, and they were falling off or jamming the conveyor belts inside the 3 PL. In defense of this 3 PL, which is otherwise a superb business, and the Chewy folks would say that this was just a mismatch. Nobody had ever figured out how to ship these kinds of products before through an E commerce channel successfully. Clearly pets.com hadn't and went bankrupt famously 12 or 13 years before. They really had an issue here, which is it was not a deal made in the spirit of partnership, meaning both sides had a 30 day out. They could renegotiate rates, which were a fee per package, shipped anytime either party wanted. Which of course was an advantage to Chewy, thinking they would drive a harder bargain as their volumes increased. But it also meant that the 3 PL could walk away the minute it got too painful. And by the way, was not particularly interested in gambling on acquiring hundreds of thousands of additional square feet of distribution center space for a bunch of guys who are running this thing on a shoestring.

Yeah, and just think about the kinds of products that I've bought from chewy.com, it's not a straightforward proposition being the fulfillment center for this kind of a business because you've got everything from small items, you know, cushy toys or whatever, to enormous bags of, you know, 50 pound bags of dog food.

So it's pretty complicated, very complicated. I mean, only 10 or 20,000 SKUs. So just to compare that to an Amazon that reportedly has 4 to 500 million SKUs on its site if you include Amazon Marketplace or Wayfair with 10 to 12 million SKUs the last time I checked. So on one hand, not a lot of products to keep track of by comparison. But on the other hand, as you say, huge variability around value, around size, around fragility. On top of that, you've got two categories that these other platforms don't face, one of which is called perishable because a lot of this food is fresh or needs to be refrigerated and will actually obsolesce on the shelf. And the other is veterinarian prescriptions. So medicated product around which there's more security and more time sensitivity in addition to all your other headaches.

Right. Can you talk a little bit more about pets.com, because you mentioned it a couple of times. It's brought up in the case. And I'm wondering, that was sort of a cautionary tale for these founders. What did they learn from that? And what did they do differently, maybe, than pets.com did?

Well, what's funny about it is what they didn't learn. The first conversation I had with Ryan, I sort of asked, at least for somebody of my age, the obvious question, which is, we all know about the pets.com wipeout. I mean, just to refresh listeners, this is a company that went public early in 2000, an inauspicious time. What we know now, with 2020 hindsight, about when the NASDAQ began its meltdown in April of that year. Its first quarterly earnings report was for Q2 of 2000. It had a top line of 8.8 million and a bottom line of 22 million in losses. Nine months later, it was in Chapter 11 bankruptcy. It was such a wonderful wipeout. And partly because those are pretty amazing numbers. I mean, they started at a $300 million valuation and of course ended at zero. And that all happened in the space of a year. But the other part of it, folks will remember that there was a $25 million television campaign to promote the talking sock puppet. And many of us at HBS have sock puppets from pets.com our offices, as humbling reminders of what it means to ride a wave in the wrong way. But in all seriousness, this created such an impact, especially on the venture community as well as founders, that nobody touched this space for over 10 years. It was like a nuclear winter of E commerce in this one sector of the online commerce universe, where just people looked at it and said, nobody can make any money there.

Wow.

So to learn from that, one issue is this issue of the pet becoming a member of the family. Humanization, the increasing price points based on human grade product premiumization. And then on top of that, the fact that at the time of pets.com, there are maybe 250 million people around the world shopping online. I mean, around the world. By the time Chewy started, there were 5 billion.

Let's talk about the four different options that Ryan and his team are considering and sort of the pros and cons of each. Maybe you can tick through those. We have three hours, right?

We have three hours. Of course, we're going to give away all the drama the next time we discuss this case in our classroom. But in a sense, it's 1 versus 3. So the issue of whether you slow down is a critical One that is, we find well worth debating and very fun to debate in the classroom. Needless to say, our MBAs, many of them are very gung ho about growth. But some of them, you know, have a sobering reaction to the fact that you've got a board that is categorically against you saying, you know, clean up the relationship, get your gross margins in shape, stop buying below cost, you got a lot to do before you should be growing at double digit or triple digit year on year rates. In any event, we normally make quick work of that first option, which is simply that if you believe scale is the way you prevail and you also believe that you've got competitors hot on your path, it's not terribly realistic from a competitive standpoint to slow down. Hence that focuses you, if you will, on doors number 2, 3 and 4. So the idea of adding a 3 PL is attractive partly because Chewy is only doing business in the eastern half of the US which is interesting. The site is clearly available across the country, anywhere in the world, but they're only taking orders from eastern half of us so they can do one to two day order fulfillment. And this is Ryan's view that if it's worth doing, it's worth doing well. So it's tempting to say if you have a 3 PL that's under stress and you want to move to a national footprint, find another three pl, see if you can build a better relationship with them, put it on the west coast, do national distribution. That's interesting. Door number three, which is don't just do that, but actually start your journey to learn how to insource logistics and fulfillment is to go build a fulfillment center somewhere. Understanding that it may take 12 to 18 months to get it up and running, you may stub your toe. It may actually be an existential risk, but at least you haven't put the entire business at risk because you've got the safety net of the existing three pl.

Interesting.

And the final option across these three growth options of the four choices we present to the students in class is simply to say if we're right about the projections, is only going to get larger. It will only become more complex with every passing day to move from an outsourcing to an insourcing arrangement for fulfillment. The time to start is actually yesterday, not tomorrow, so let's do it now. And that would mean severing relationship with existing 3 PL. Enormous risk in doing that because you're not only doing a switchover which is risky enough to another facility, but you're doing it to a facility that you haven't even built yet. And that one is a really tough one because industry experts are telling Ryan and his team that it's a $10 million capital expenditure to stand one of these things up, that it's then going to cost you the better part of a couple million dollars a month in operating expenses to run, and that all the best consultants they could tap in the industry are essentially saying you can't just flip a switch. It really will be best case 12 to 18 months before you're actually up and running.

Yeah. I'm curious as to how Amazon started on this. Did they use a 3 PL when they first began selling books? How did they do it?

It's a great question. Arguably the Seattle location was part of the secret to answering that question. Amazon in the beginning, located in Seattle because quite proximal to their own modest warehouse, was a large fulfillment center operated by the largest book distributor in the United States called Ingram. And so Amazon in the early days kept the top 11 to 12,000 fastest moving titles in its own warehouse, but nonetheless put 7 million titles on the site with the understanding that they could send a truck driving over to Ingram and pick up any one of those titles at any time. Obviously not all 7 million. Sure. But the point being that Ingram did have hundreds of thousands of titles sitting in that facility and Amazon essentially could do reasonable order fulfillment time without actually taking title to owning or storing all of that merchandise.

Yeah. What a great strategy.

Great strategy. To your point, avoided the vagaries of three pl, worked with a world class fulfillment center that happened to have comprehensive inventory and they didn't have to take title and they didn't have to manage it.

Yeah. I'm also curious about how involved the board gets into these conversations. This is maybe more general for entrepreneurs. They've all got boards that they have to answer to and advisors. The board here clearly had some strong about what Ryan and the team should do. Is Ryan sort of, does he have to listen to them or is this his call? Like how does that play out?

Nobody was against this idea of growth, but they were looking at this business and saying it is not making a lot of economic sense right now. Everyone was waking up in the middle of the night thinking about pets.com because nobody wanted to see a wipeout like that. And the presenting problem was was there some kind of middle area? Goldilocks answer. Meaning you don't stop growing, but could you slow down growth enough to improve the existing relationship and de risk the fulfillment situation to some Extent versus what Ryan and his co founders wanted, which was to swing for the fences because they viewed any kind of step back from this meteoric double, triple digit growth as essentially a concession, as waving the white flag and acknowledging that they would not win in the category.

So I know you wrote a B case. Are you allowed to reveal what decision they made? Can we let our listeners in on this?

I think we can. It's not easily googleable, but interestingly, Ryan and the team prevailed in the boardroom. And by that I mean that, you know, everyone ultimately agreed to disagree and did commit to a path forward. And the path forward was to the riskiest of all of those options. And that was door number four, which is essentially to sever relations with existing three PL and go all in on fulfillment. That turned into exactly the kind of nightmare situation that you and our listeners might expect. I mean, the first thing that happened is as they were standing up the facility in the same town in eastern Pennsylvania where the 3PL was located, it did not take long for the quote unquote secret to get out that Chuy was establishing its own fulfillment center facility. When that happened, the company they were dealing with, their current 3 PL partner, said, Gee, we have no contractual relationship. We're charging $3 a package shipped. How about we go to nine?

Oh my gosh.

So they tripled prices on Chewy while Chewy was sitting there saying, we can't take 12 to 18 months to get this thing up and running. But the logic was the one we talked about, which is, if you're going to do this and if it's a core part of your proposition, and again, it comes back to what you were talking about earlier, Brian, if you believe that part of getting it right for customers is establishing this human connection through the call centers with these handwritten notes, sending flowers when a pet passes away, sending beautiful fine arts oil portraits, which Chewy does as well. For people who send JPEGs or images into the call centers. After a wonderful chit chat with somebody down in Fort Lauderdale.

Amazing.

It's hard to argue that a core element of getting it right for customers isn't, for lack of a better term, the out of the box experience, meaning the whole issue of does the product arrive in time? Is it nicely packaged? Is the order accurate? Does your dog or cat or goldfish like what they shipped? So the view was that this actually was not peripheral, this was core. And if it was core from a competitive differentiation perspective, then it was something they had to ultimately own and operate the other. Is that all Four of these choices are awful. Each one, as we said earlier, has just powerful arguments against it. And so this idea that you're in a situation where you want to save the company, you've got some pretty well defined paths forward, but no one of them is attractive. What puts me in mind, I often say this to the students, is that in the normal world of business, we talk about necessity being the mother of invention. But in the startup world, and this is an extreme case, you could argue the opposite is true. True invention is the mother of necessity. The fact that they'd created a business that had gone from just a few million dollars in 2012 to a business that ultimately closed out 2013, the year of the case, we're in December of 2013 at 73 million. That's invention. Something was working in a way that it wasn't working for anyone else who'd ever touched this sector, either brick and mortar. Or online. But that created this, in a sense, existential moment in which the demand would either crush or you could capitalize on it intelligently and save and make the business.

Yeah, not for the faint of heart, those kinds of decisions.

Not at all. Yeah, not at all.

Well, Jeffrey, this has been a great conversation as usual, so I expected nothing less than a great conversation with you. But I'll give you an opportunity to sort of finish up by telling our listeners if there's one thing you want them to remember about the chewy.com case, what is it?

Let me start by answering that with just a quick update on where they are.

Great.

This is a business, of course, that took a massive gamble. But as we said, any one of these choices would have been huge risk. They ultimately sold the company for $3.35 billion to one of the big boxes, Petsmart. It was an all cash transaction that represented the largest US e commerce exit in history at the time. Petsmart, laboring under enormous debt from an LBO they had done with a New York and London based private equity firm, needed some way to get their hands on cash. And so it was a bit of a Hail Mary. They spun out Chewy in an IPO just a few years ago. Chewy went on to become at the IPO a $17 billion public company. With the uplift during COVID it reached 40 and 50 billion. And even now, with the down markets we're living through today, it's still in the range of 15 to 20 where it IPO'd. So just to reassure listeners that there are times when people do crazy stuff as entrepreneurs, when if they're smart and savvy and a bit like things can really work out. But for us, from the standpoint of lessons for our students, one of the big things that scaling startups face is that moment when you make the big bets converting what are variable costs into fixed costs, which is ultimately what this fulfillment decision is all about. And those are very risky decisions for the simple reason that if you don't make them and you outstrip your partners, you can't deliver for customers. But if you do insource and convert variable to fixed and you don't deliver on your revenue targets, you crater the venture.

Yeah.

So back to what you said, not for the faint of heart.

That was Harvard Business School professor Jeffrey Rayport in conversation with Brian Kenny on Cold Call. We'll be back next Wednesday with another handpicked conversation about business strategies from Harvard Business Review. If you found this episode helpful, share it with your friends and colleagues and follow our show on Apple Podcasts, Spotify, or wherever you get your podcasts. While you're there, be sure to leave us a review. And when you're ready for more podcasts, articles, case studies, books and videos with the world's top business and management experts, you'll find it all@hbr.org this episode was produced by Robyn Pashas, Craig McDonald and me, Hannah Bates. Kurt Nickish is our editor. Special thanks to Ian Fox, Maureen Hoch, Erica Truxler, Ramsey Kabaz, Nicole Smith, Ann Bartholomew and you, our listener. See you next week.

Sam.