Case Study: Should a Direct-to-Consumer Company Start Selling on Amazon?

Sitting in his office, Mark Ellinas frowned at his computer screen. It was filled with row after row of electric bikes, from expensive models to cheap knockoffs that seemed held together by spit and a prayer. Though they varied in style and price, the bikes did have one thing in common: where they were being sold. The website he was looking at, flush with options, was Amazon. As the CMO of PedalSpark, a small maker of high-end electric bicycles, Mark was considering strategies for selling the company’s new ride. The market for electric bikes had exploded in the past few years,1 especially in China, and it showed no signs of slowing down. PedalSpark’s signature bike, a $4,000 luxury model available only through the company’s website, was selling well and had been named to a few “best e-bike” lists. Now PedalSpark was about to introduce a cheaper, entry-level model, which it hoped would have broader appeal. The bike was targeted at price-sensitive riders, people who were willing to trade higher battery life and motor power for a lower price tag. CASE STUDY CLASSROOM NOTES 1 E-bike sales are projected to reach 40 million units by 2023; more than 34 million of those will be sold in China. Two years ago PedalSpark had hired Mark away from his marketing position at a children’s bicycle maker. That company had sold exclusively on its own site, and Mark’s expertise had served PedalSpark well with its first product. He was excited by the challenge of selling the new bike in an increasingly crowded market, but the question was how to do it. His two direct reports were split. Gideon Bear, the sales manager, tended to favor aggressive approaches. He wanted to sell the new model on Amazon, which had, as he’d put it, “a few more customers than our site.” But Tamar Nourse, the product manager who’d recently come on board, was worried about whether the bike would stand out on Amazon. She thought that keeping the new model on PedalSpark’s site, where their team could control the entire sales process, would be better over the long term. Bzzt. Mark glanced at his phone and saw a text from the CEO: Where are we on the online channel strategy? Looking forward to your presentation. The new model was almost ready, and the CEO wanted a decision soon. With the presentation scheduled in two days, Mark still had some time to think—but not much. Giving Information to the Enemy Mark closed his laptop and walked down the hall to Tamar’s office. He knocked on the open door. “Hey, got a minute?” Tamar looked up and adjusted her thick-rimmed glasses. “Hi, Mark. What’s up?” He sat down across from her. “So, about the bike. In the meetings with Gideon it feels like you’ve been holding something back. We have to make a decision, so I need you to tell me what you aren’t telling me.” She took a deep breath. “Mark, I’m still new here, and I don’t want to rock the boat. But I really think selling on Amazon would be a terrible move for us.” “Why, though?” “The day we put the bike on sale, Amazon will start vacuuming up information about our customers, our margins, and the market’s potential. If it ever decides to get into the e-bike business, we’ll have hand-delivered all the data it needs to squash us.” “I know that worrying is part of your job, but is it possible you’re being a little paranoid here?”2 2 Amazon has a huge advantage over the merchants on its site. What else should Mark and PedalSpark’s leaders think about to improve their chances of succeeding? “You should ask my B-school classmate Marta.” “Who is she?” “A few years ago she was the founder and CEO of a successful start-up. She’d had an idea for a new kind of tablet stand. She spent a year developing a prototype and finding a manufacturer in China that would work with her. Then she started selling on Amazon. Now she’s the former CEO of a company that doesn’t exist anymore.” “Wow. What happened?” “For about a year the tablet stand got great reviews and sold well at $40 each. During the back-to-school season, she was moving a few thousand a month. Then a bunch of copycat products started popping up. She had to fight them off as best she could. She complained to Amazon, but it didn’t do anything, of course. Then AmazonBasics debuted its new tablet stand.3 It was a lot like hers, though different enough to avoid a lawsuit. It was also half the price.” 3 Amazon has about 130 private labels so far, in areas ranging from simple electronics to clothing to pet food to furniture. “E-bikes are a lot more complex than tablet stands, though. What are the chances Amazon will make one of its own?” Tamar’s lips curled into a small smile. “I don’t know, but if we went head-to-head against Jeff Bezos, would you put your money on us? Amazon’s private-label products are projected to hit $25 billion in sales by 2022.” Mark shuddered. “A dark thought to have before lunch. How do you figure our chances against the existing competition?” “We do have great bikes, but quality isn’t enough on Amazon. Whatever your product is, there’s always a cheaper version, and usually that’s the one people buy. It’s a never-ending, anything-goes price war there. I’m guessing that isn’t what we want people to associate with our brand.”4 4 If PedalSpark has to compete on price, what might be the consequences for its brand image? Nodding slowly, the CMO rubbed his chin. “Good point, and I don’t disagree. Gideon is pretty keen on the Amazon idea, though.” Tamar adjusted her glasses again. “I get why—more customers and more visibility. That may help us sell bikes in the short term, but what about the long term?5 If people buy the new model on Amazon, will they be loyal to the maker or to where they bought it? We built the PedalSpark brand by selling the luxury bike on our website. Why try to fix what’s already working?” 5 As the e-bike market continues to grow and mature, what factors will determine which companies succeed and which don’t? Trying Something New That afternoon, Mark asked Gideon to meet him in the cafeteria for coffee. The sales manager poured milk into his steaming cup and swirled it around with a straw. “Amazon, Mark. You know what I think. What are you thinking?” “Undecided. There’s a lot of risk in selling the bike there, but a lot of upside, too.” “Yes! I’m glad you see that. Amazon Prime has over 100 million members,6 and it’s growing. Imagine the sales if a fraction of them ordered the new bike—and imagine how many of them will if two-day delivery is available. Someone gets excited about e-bikes on a Wednesday, and by Friday she has one of her own to ride. The possibilities are endless.” 6 Of the U.S. consumers who make $150,000 or more, 70% are Amazon Prime members. “It’s fun to daydream about, Gideon, but are we set up to handle higher volume and a shorter fulfillment window? Orders that come through our site have a shipping time of two weeks. I’m nervous about promising something we can’t deliver—and to a bunch of new customers, no less.” “But that’s the beauty of Amazon,” Gideon said, his voice rising in excitement. “We have options. I know I’m telling you your job right now, but we can sell product to Amazon for it to resell, or sell the bikes ourselves and let Amazon handle the warehousing and shipping, or list them on Amazon and ship them on our own. You’re always talking about the value of running small, controlled experiments, so let’s try one and see what happens. If it doesn’t work, we’ll switch tactics and adapt as we learn.” He grinned. “Everyone in this company agrees we have a great new product. All I want is to get it to as many people as possible.” “There are three options, yes, but they don’t give us a lot of wiggle room if things go badly. We may be able to play with the bike’s price a bit, but we can’t lower it that much or we won’t make any money—and it could make us look cheap, too. I do think a higher price point is fair for the bike we’re selling. Even luxury brands that sell on Amazon today hesitated about it for a long time, and it would be a good idea for us to think about why that is. The jury is still out on whether they benefit by being on Amazon.” “You know who sells on Amazon? Apple. Versace. Rolex. Jimmy Choo, Mark—Jimmy Choo. And more will follow. Whichever companies don’t will be on the wrong side of retail history.” “We aren’t Versace, Gideon. Besides, a lot of those brands sell a very small subset of their products on Amazon—and usually not their flagship ones. They save those for their own sites or stores, where they can control the buying experience. We’re trying to raise our profile as a high-end brand, right? How would we look if we were one of dozens of e-bikes in Amazon’s listings?”7 7 How might selling on Amazon help—or hurt—PedalSpark’s image as a high-end brand? “Sure, but we already have the luxury bike selling well on our site. I agree, we shouldn’t change anything there. But the new bike is for everyone. And everyone is on Amazon.”8 8 When looking for a product online, more than half of consumers (54%) start with Amazon. Mark took a sip of coffee, thinking. “Look, I get it, you have some concerns,” Gideon continued, “so let’s talk numbers. Based on what our competition is doing, I figure if we put the new bike on Amazon, we can reasonably expect to sell 10,000 units a year.” “At what price point?” “$899. That’s a little higher than we’ve been talking about, but it gives us some room to go lower if we need to.” “And what are the latest numbers for luxury bike sales on our site?” “Last year we sold 2,000 units at $4,000 apiece. Remember, the new bike won’t be only on Amazon. We’ll sell it on our site, too.” Mark scratched his head. “What we really need is a way to quantify the risk that Amazon will enter the e-bike market. It would make this so much easier.” “That’s the big mystery. Amazon will have all the consumer data, and we’ll have very little of it. But look at it this way—there are already a lot of e-bikes on Amazon, so they’re already watching the market. Even if they do make their own bike, that could be years away. We might as well find new customers while we can. People can’t buy our bikes if they don’t know about them.” Mark stared at Gideon for a long moment. “Let me ask you something. How are you so sure about all this?” Gideon laughed. “In my moments of doubt, I think of Instant Pot. It’s a quality appliance—not quite luxury, but good—that has a cult following and that made its name on Amazon. At one point, 90% of its sales were from there. Do you know how many Instant Pots were sold on Prime Day this year?” “No, but I’m a little surprised you do.” “I cook a lot. The number, Mark, is 300,000. In just 36 hours. I think we could be the Instant Pot of e-bikes.” The CMO stirred his coffee. “You may be excitable, but I’ll admit it’s kind of contagious. I just can’t shake the feeling that once we open the door to Amazon, there will be no closing it.” Gideon held up his coffee for a toast. “To opening the door—just a crack—and seeing what’s behind it.” Searching for Answers Back in his office at the end of the day, Mark was staring at his computer again. Tamar and Gideon seemed so sure of what to do, but the CMO was struggling to make up his mind. The screen of his laptop still showed the Amazon site, with its rows of e-bikes. Sighing, Mark opened Google and typed “What are the dangers of selling on Amazon?” into the search bar. The query returned almost 250 million results. “Hard to tell whether there are more horror stories or more success stories,” he muttered. “Well, this bike isn’t going to sell itself. I have to decide something, one way or another.” Question: Should PedalSpark sell its new, lower-cost bike through Amazon? The Experts Respond Stephan Aarstol is the founder and CEO of Tower Paddle Boards and the No Middleman Project. PedalSpark should sell on Amazon for as long as it makes sense. Amazon is using PedalSpark and other brands until it doesn’t need them anymore, so PedalSpark should do the same with Amazon. Because the retail landscape is constantly changing, selling there in 2019 is very different from selling there two years ago. And in two years it will be different again. If Mark and his team determine that margins on sales through Amazon now are good enough, they might as well test out the marketplace. Amazon does take a healthy cut of transactions and requires brands to pay just to be seen, so those margins could be slim. But the extra exposure might make up for it. Mark will need to keep a careful eye on how his niche evolves on Amazon and how margins change over time, however. When—not if—selling on Amazon isn’t profitable, he can pull the new e-bike from the platform and offer it exclusively on PedalSpark’s site. No matter what PedalSpark does, Amazon will get a ton of data on e-bikes and will jump in with its own products eventually. The company is known for its AmazonBasics label but has other, higher-tech house brands, and most consumers buy them without realizing it. In fact, perhaps PedalSpark should try to sell its original luxury bike on Amazon instead. The margins on the new, cheaper bike will inevitably be lower, so why not put the established, high-end offering on Amazon and introduce new customers to it? That may cannibalize some sales, but so will other comparable offerings on Amazon, and in the meantime Amazon is an exponentially bigger market. Why not put the high-end offering on Amazon instead? If the cheaper bike goes on Amazon first, that’s what customers will associate with the firm’s brand. Is that really what Mark and the CEO want? They need to think about the long term. They should use Amazon primarily to build the PedalSpark brand, with a view to driving customers to their firm’s own site for future sales. Wisely using Amazon is not just about increasing short-term transactional volume. This case study is loosely based on my experience as the founder of Tower Paddle Boards, a start-up that’s one of the biggest success stories from the TV show Shark Tank. But when we started selling on Amazon, back in 2012, the market was new, there weren’t many competitors, advertising wasn’t required for visibility, and the margins on the site were much higher. It was just easier to succeed. As Amazon evolves, it looks more and more like an online convenience store with traditional retail markups. It has everything—but it’s getting too crowded. I suspect that many of the best brands will start to pull their products from the platform and return primarily to direct sales, as we did recently. Long term, the best value for consumers will be buying from direct-to-consumer-only brands. We’re trying new ways of advertising and selling, and we’re mapping out a future in which we’ll be fine without Amazon. If you have a great product and know how to sell it, customers will find you, wherever you sell. Gil Efrati is the chief marketing officer at Nectar Sleep. Before thinking about selling on Amazon, PedalSpark has to build a brand customers recognize. If it doesn’t do that first, its e-bikes may get lost in a sea of similar products on the site. When you’re a young consumer-product maker, success depends on differentiation—on the way you stand out. On its own site, PedalSpark controls the user experience and owns the sales process; it knows its customers, can promote loyalty, and can create scarcity by limiting the ways to get its product. On Amazon that’s all nearly impossible to do. These days the platform is essentially the product search engine. Shoppers love it because it lets them comparison-shop and has an easy checkout process, responsive customer service, and speedy, low-cost shipping. But they’re loyal to Amazon, not necessarily the brands they’re buying. And Amazon keeps their data and controls the relationship; the brands know very little about these customers and have no way to contact them to upsell them. Gideon is right that Amazon offers access to a huge new market that cannot be reached any other way. There’s no question it can be an extremely effective distribution channel for established brands. But PedalSpark must build its name so that it can operate from a position of strength. It needs customers to be searching not just for “e-bikes” but for “PedalSpark e-bikes.” High-end brands can sell on Amazon because people are looking for an iPhone or Versace sunglasses, not merely browsing categories. Start-ups such as Warby Parker, Bonobos, and BarkBox have had more success with direct-to-consumer sales. My company, Nectar Sleep, has begun selling some of our high-quality mattresses on Amazon, but we did so only once our brand was strong enough to succeed there. People do search for our products and give us good reviews, so we feel we can withstand the competition, even from Amazon, which sells mattresses through a private label. We work to offer a better customer experience on our own site, where the vast majority of our sales happen. When people buy from us directly, they get a 365-day trial, a lifetime warranty, and other benefits. Most important, we know who they are and can direct them to other products they’ll like. Customers are loyal to Amazon, not the brands they’re buying on it. It sounds as if PedalSpark is doing a good job with its own online channel, but there’s always room to improve, whether by enhancing the user experience, offering targeted discounts, creating a bigger social-media presence, or doing guerrilla marketing. If the goal is quick sales that boost cash flow, selling on Amazon will do the trick. But it won’t necessarily lead to long-term growth and profitability. History shows that no strong direct-to-consumer brand has emerged by selling on Amazon. PedalSpark would be wise to create a loyal following and brand equity for its e-bikes before betting big on the platform. HBR’s fictionalized case studies present problems faced by leaders in real companies and offer solutions from experts. This one is based on the HBS Case Study “Selling on Amazon at Tower Paddle Boards” (case no. 517047-HCC-ENG), by Thales S. Teixeira and David Lopez-Lengowski.

Case Study: Should a Direct-to-Consumer Company Start Selling on Amazon?

Partner Center