Thales Teixeira, associate professor at Harvard Business School, believes many startups fail precisely because they try to emulate successful disruptive businesses. He says by focusing too early on technology and scale, entrepreneurs lose out on the learning that comes from serving initial customers with an imperfect product. He shares how Airbnb, Uber, Etsy, and Netflix approached their first 1,000 customers very differently, helping to explain why they have millions of customers today.
CURT NICKISCH: I want to talk to you about these early-stage companies because you focus on that in the book, taking your business from zero to 1,000 customers. Clearly, that’s a different stage, but why is it so unique?
THALES TEIXEIRA: So, in my experience at Harvard, I teach courses on e-commerce and I teach courses on digital marketing and quite a few of my students in the past few years have been launching their own businesses and they start off the process of launching by trying to emulate companies that are successful and are similar to them.
CURT NICKISCH: This is where you get everybody saying the Netflix of this and the Uber of that.
THALES TEIXEIRA: Yeah, so they already emulate it in the way they think about the business, but they also emulate it in the way they try to acquire new customers. And I say, “Wait a minute. Why – you’re a startup, you’re trying to acquire your first 10 customers. Why are you looking at Uber who’s already had millions of customers and is trying to acquire them?”
If you’re trying to learn how to walk and you look at Usain Bolt, the fastest man on the planet and try to observe how he’s running and say, “I’m going to do the same thing.” It’s not going to work. Why? Because you’re at a completely different phase in your life.
And so, unfortunately for these entrepreneurs, there are not many articles or explanations of Uber, Airbnb, and Amazon, what did they do in the early days? How did they acquire their first 100 customers or 10 customers? What they did, in the beginning, Airbnb, Uber, Etsy, and all these other big companies is not what they’re doing now that they’re huge.
CURT NICKISCH: You’re looking basically at Uber’s baby steps.
THALES TEIXEIRA: That’s right.
CURT NICKISCH: To figure out why they’re running as fast as they are today. What is different about those first steps and how are early customers different?
THALES TEIXEIRA: So, you’re a startup. You don’t have a lot of money. You don’t have all these resources. Therefore, the chances that you’re going to build a product or service that is so much better than what’s available out there in the market today is very, very slim.
And so, when you acquire those first few customers, that is practically an aberration. It shouldn’t have happened. You should really keep asking yourself, “why did they buy from me and not the alternative?” You should make it a puzzle.
In my experience, these startups, go too much in the process of optimizing to get as many customers as possible at the lowest cost possible. And they forget that what they’re trying to do is understand what should they build for their customers and why are customers even interested in their products?
CURT NICKISCH: I mean it’s understandable why they do that. If you don’t have a lot of resources you want to keep costs low, but that takes you down a different path of thinking that you essentially, end up learning the wrong things and developing the wrong skills at that stage.
THALES TEIXEIRA: That’s right. And the alternative to acquiring customers at a low cost is not acquiring customers at a high cost. In the beginning, the alternative is not acquiring customers at all.
In the beginning, Uber did not have the great experience that it has today. You would download an app and you would actually have to send a text message or call somebody and there would be an Uber employee that would receive your message or your phone call and would frantically try to call other black car companies – that’s how they started – and try to get somebody to go pick you up wherever you were. And so, it wasn’t this automatic process that it is today.
The same thing with Airbnb. Airbnb was actually you were renting an airbed which is an inflatable mattress and somebody would make you a hot meal once you woke up. Are those 10 times better products than what’s available out in the market today? There are so many good hotels. They were not. So, why? Why would anybody go through this process?
Well, the secret is these businesses, in the beginning, looking for pockets of demand, of consumers that weren’t satisfied, weren’t being satisfied by the traditional players. So, Airbnb launched at the national conventions of political parties, in San Francisco, and in Austin, they did a South by Southwest.
And during those times these cities had 100% occupancy in hotels. And so, if you were one of the last ones to go there you just basically had no place to stay. And so, when Airbnb came and said, “Hey, stay in somebody’s house and pay 30, 40, 50 bucks per night,” people were like either I do that or I don’t have a place to stay.
The same thing Uber – Uber launched in places and in times when people could not get cabs. Think about after a major sports event, or a concert. People, there were thousands of people coming out needing cabs at the same time. And so, Uber made sure to be in these places where the alternative was nothing. And that’s how it acquired its first customers.
CURT NICKISCH: So, in those cases, they’re not even competing with anybody at that time.
THALES TEIXEIRA: Exactly. And that’s the point. Figure out how you can compete with no other options. You need time and resources to build a better product. And in the process, you need to acquire these first few customers that are going to help you in some regards.
First, they don’t have options. Second, customers do more for your business in the beginning than just pay you, just give you a check. Customers can do so many things. Number one, customers can talk to other acquaintances about your product and that’s much of how both Etsy and Uber came about because people used them and they liked them and they told other people about it. So word of mouth was important.
Second, these early customers, there’s a reason why they’re going for you because they’re not fully satisfied, or they’re happy to kind of take a chance on a new startup. So, you should make sure that you take care of them and help them help themselves. Why are they doing that? They’re interested in talking to their acquaintances about new startups. New apps. New things like that. So, support them and give them the resources.
Third, these consumers can actually provide you with R&D that you traditionally would need to test and evaluate, get consumer groups, and focus groups, and pay for that. They’re willing to tell you for free what is working, and what is not working in their business.
There’s even a startup that I advise. It’s a business-to-business application. So, instead of hiring people to go on the phone to do product support and help when their customers call in and say “I don’t know how to do this, I don’t know how to do that.” They actually gave these consumers the task of teaching the new consumers how to operate the software.
So, all of these things consumers are willing to do, and the alternative would be to hire people to do marketing, do R&D, do product development, do product testing, to do consumer support.
CURT NICKISCH: You haven’t mentioned technology, or hiring coders. You haven’t mentioned a lot of the things that we associate with big, new technology, disruptive companies. Are you going to get to that or is this something that’s intentionally missing?
THALES TEIXEIRA: So, this is intentionally missing because this is intentionally not part of the success stories of startups in their early days. Uber did not have any innovative technology in its early days. Those drivers actually had GPS connected phones, so for the first time ever you would see a car coming closer to you. But that was it.
Airbnb had no new technologies. Even Amazon in the beginning. They were selling books. No new technologies. Etsy, no new technologies. Netflix had no new technologies. They were just buying DVDs and shipping them from different shipping facilities.
And so, the important thing that a tech entrepreneur should see is that it’s this idea that we commonly refer to as the iceberg. Only the tip is the technology, but under the technology, there’s a lot of operational excellence in all these companies. These startups, in the beginning, had to just go outside of the internet and recruit customers knocking on doors.
Etsy decided to actually go into all these fairs across the United States, having people travel to all these craft fairs and recruit sellers because the sellers would eventually recruit their own buyers to the website instead of just using digital marketing. The technology eventually becomes important in differentiating factors. But in the early days, it’s definitely not true.
CURT NICKISCH: Yeah. This makes me think of Pinterest also which really incubated its early customers by being really judicious about who they invited onto the platform at first. So, they really focused on designers and photographers, and they tried to create a culture that eventually everybody else would emulate when they weren’t able to manage every single person on there as they did at the beginning. But it just sort of helped set the stage for what they were trying to do.
THALES TEIXEIRA: And it helped set the tone for quality. So, there’s a great story on Airbnb that it was doing well in many cities in the U.S., but it was doing very poorly in New York City. And the founders of Airbnb went there. First, they went and stayed in a few Airbnb rentals – a few private homes – to try to understand what was going on and it turns out the quality of the homes, and the apartments were pretty good, but the quality of the images in the website was really bad.
So, they decided to instead of trying to convince everybody to take better pictures, they just rented a very expensive camera and went from apartment to apartment in Manhattan and would say, “we’re going to take the pictures for you” and they would take professional quality pictures instead of the cell phone pictures that were being uploaded.
And suddenly people started seeing this in a better light and that created the standard for all of the other ones. So, this comes back to this – do things that don’t necessarily scale. Scale is a problem when you’re growing fast. If you’re small and you don’t get to that phase, there’s no point in trying to do things at scale if you’re going to die beforehand.
CURT NICKISCH: One of the common things that any founder is told who has an idea is that you need to partner with your CTO early on. That’s maybe your co-founder is going to be a technical person. And so, I’m just wondering if that is necessary? Are these startups that start out by building an app and really focusing heavily on the technology early – are they going about that wrong?
THALES TEIXEIRA: Well, I think that it depends actually on what type of business we’re talking about. If it is a business in which the value to the consumer will be delivered through the technology primarily, you need to build something. You need to build the infrastructure and that a CTO will do. But if the infrastructure to deliver value to consumers is not primarily through technology, through an app, through a website then probably you can get away early on without having to hire a partner with a – you know, a CTO that will build a technology.
So, to give two examples. So, all of these startups right now, there’s this craze of direct-to-consumer startups that are selling anything that you can probably see in your home, that you would normally buy in a grocery store, or in a supermarket. Now there’s a startup, they’re just trying to make a better product and deliver it for you. Toothbrushes. There are beauty products. Pet care. All of these things. And mattresses.
The technology’s not the primary driver of the experience for these products, it’s the products themselves. And you can easily create a website and offer that product. And so, having a CTO early on is not kind of the main driver of your early customer adoption.
Whereas obviously if you’re building kind of a business to business software, then the way you’re delivering it is through software and obviously having a CTO will be of primary importance. So, it depends on the type of business. The question that entrepreneurs should be asking is, what is the primary driver of value to the customers that I will deliver and how technology play a role in that, how big or how small?
CURT NICKISCH: Yeah. How do you think Airbnb answered that question early on?
THALES TEIXEIRA: Airbnb had one engineer for a long time.
CURT NICKISCH: Which was one of the founders, yeah.
THALES TEIXEIRA: Founders, yeah. And that’s it. Right? So, Airbnb’s whole purpose was to reduce the– or increase the utilization of an expensive asset which is a private home, or a second home, or summer home, or something like that.
And you don’t really need that much technology. When they launched, they faked it a little bit. Instead of just showing their inventory they hid it. And so, when you would go online they would just give an example of a few places and probably that’s all they had. And you would go and say “OK. I need to be in New York City on the 17th, or the 19th, and I want an apartment.”
Well, you can’t really choose from all the available inventory because there was none. And so, they just said, “you fill out that information and then somebody will come back to you with a few options in 24 hours or so.”
So, the experience was not perfect – like today you can just look at the options and even book it online and have it guaranteed. This is way later on. In the beginning, you have to be humble about what you have and what you can offer. And so, if you don’t have it you have to do a little bit fake it until you make it.
So, don’t show your weaknesses. Show the availability or possibility. And the first customers that went to Airbnb were the ones that were like “Fine. I don’t need it, I’m not a business person. I don’t need to make sure I have a guaranteed booked in the next hour or so, but I’m interested in staying in a house next time I visit San Francisco.”
And those are the customers – were the ones that they got. It wasn’t those like the mass population that requires additional things. So, that’s what’s important here.
CURT NICKISCH: Yeah. This is making me think of enzymes in chemical reactions where the reaction would take place anyway with a lot of time and maybe with a lot of energy. And enzymes essentially lower the amount of energy for such a reaction to happen and they’re–
THALES TEIXEIRA: That’s a great point. So, we could say that technology is the enzyme of startups. Technology is never the driver of digital disruption, in my experience, from my research. Technology is always the enabler or the way to speed up the process.
But it speeds up a process once you know what process you want to speed up. If you speed up a bad process, you’re going to just drive through the cliff with your startup faster. So, you should always think about — first — kind of figuring out what product or experience you’re going to build and deliver and then acquire your customers that are willing to coexist to purchase these products even if they’re not perfect.
And there are people out there. There are people that are trying new things. There are people out there that prefer to give their business to a startup than a big company. There are people out there that they are expecting to have a relationship with other companies. There are people out there that have all these other different needs – that they don’t want to go to the mass of the market and the standard products and services available. And so, catering to those is very important. And you will find those not by plastering ads on Facebook or on Google at very expensive costs.
CURT NICKISCH: How do you know when you are done with the baby steps? How do you know when you’ve learned what you needed to from that first thousand customer phase and that you need to move on and start scaling?
THALES TEIXEIRA: The key metric that changes from baby steps to learning how to walk efficiently and run fast, is in the baby steps your goal is to learn, not to do fast. When you’re running your goal is to do fast, not learn. You should have learned way before.
And so that is the key insight to know that in the beginning, the goal for every new customer that you acquire is: what did we learn? What did we learn by acquiring this customer and what did we learn by servicing this customer that we can use in the next customer and then in the next customer and again, and again, and again?
As this cycle unfolds of learning how to acquire better and learning how to deliver the service or product better, over time naturally startup entrepreneurs start to reduce their amount of learning. They, everything that they learned they already knew, so there’s no new learning.
And you start amassing a group of customers that starts to get bigger and bigger. That is the point that you should start switching from trying to learn as much as possible and disregarding to some extent, doing it efficiently into an efficiency mode, into scaling, doing it faster and faster. Getting more and more customers for each dollar you spent and being more effective. The idea is that you stop learning and then you learn this learning to optimize yourself. That is the transition point to be able to kind of switch from learning to doing it better.
CURT NICKISCH: You clearly have a sense that a lot of startups fail because they don’t master this first phase well. So, which do you think is harder? This first phase or the scaling part? Where do you think more startups fail?
THALES TEIXEIRA: Well so clearly research has shown that more startups fail in the early phases. The vast majority of startups – both tech startups and non-tech startups, small businesses – they fail in the first few years. One of the biggest reasons that they fail is actually they don’t get enough customers to pay for the cost of their maintenance or their survival. And that’s one of the main drivers.
And so, obviously, the answer to not failing as much, or not have a huge likelihood of failing as a startup is really to make sure that you acquire enough customers to sustain your business. And in that regard, this is a little bit of a shift away from what others have advocated for. There’s this idea that you should build your best product possible. You should get capital. You should grow. You could scale. You should create network effects. And you should use technology and invest and hire technical people.
All of these things are overwhelming to startup entrepreneurs and they basically can’t do them all at the same time. And so, what I advocate is: these are all important elements for later on in the phase of a company for the scale phase. But in the early phase of a company, acquiring customers is your main job, your only job and all that you have to focus on.
And the way to acquire customers if you’re going to build better products, or you’re going to create a better relationship, or do whatever it takes. You do that with the goal of acquiring enough customers to be able to just survive another day.
CURT NICKISCH: That’s Thales Teixeira. He’s an associate professor at Harvard Business School. He’s also the author of the book Unlocking the Customer Value Chain: How Decoupling Drives Customer Disruption.
The article first appeared in The Harvard Business Review