I present five models to define your product strategy then bring these models to life with a mock 2020 articulation of Netflix's current product strategy. I also demonstrate how product strategy helps day-to-day decision-making with two modern-day ""What would you do?"" Netflix cases: 1) Should Netflix enable custom playback speed, and 2) Should Netflix launch ""Netflix Party"", enabling worldwide members to enjoy shared, movie-watching experiences?
Gibson Biddle:
Hello, Product-Led Summit. My name is Gibson Biddle and I'm here to talk today about Netflix's 2020 product strategy. I'm going to talk a fair amount about retention and give you a glimpse of today's current Netflix product strategy.
Gibson Biddle:
I want to begin by talking a little bit about my early career. The first hit software ever built was called Elmo's Preschool. This is in the nineties. I was the co-founder of a kids software startup called Creative Wonders. This is the same year that Oprah Winfrey threw Tickle Me Elmos onto the stage and that beget the hit. And we went on to sell that company, Creative Wonders, to Kevin O'Leary, Mr. Wonderful, the Shark Tank, and he was the CEO of The Learning Company, and then helped The Learning Company to grow, making Reader Rabbit and Oregon Trail software, and then eventually sold that company to Mattel for over $4 billion, a big deal at the time.
Gibson Biddle:
But, in fact, this turned out to be a celebrated disaster because a year after the acquisition, The Learning Company was spit back out of Mattel, and as it turned out, it was one of the worst acquisitions of all times, a loss of $3.6 billion. And the key learning for me was I had built up software. In perfect hindsight, I had failed to build a hard to copy advantage, and so other companies were able to copy and emulate what we were doing, and that's why the valuation fell. And it was a big learning for me about the importance of building hard to copy stuff.
Gibson Biddle:
This is Reed Hastings. He's the CEO of Netflix. Most folks may not know, or remember, that Netflix started as a DVD by mail company. So, I joined Netflix in 2005, and this is when I learned a lot about product strategy and really this marks the back half of my career, where it's been much more successful than the first half. So my background, I joined Netflix as a VP of product in 2005, and by 2010 I moved on to my next startup, which is Chegg, it's a textbook rental and homework help company that saves students billions of dollars renting instead of buying, and it went public in 2014. The last five years, I've been very focused as a speaker, as a teacher. I teach both at Stanford and INSEAD in Europe. I'm also an advisor and just love to teach.
Gibson Biddle:
So, there's really three chapters to my story today. The first chapter, I'm going to introduce a few product strategy frameworks and I'll go in depth on one of them. Then I'm going to take those tools, models, and frameworks and say, this is how they would be applied if I were running product at Netflix today. Then, I've got two juicy, "what would you do?" cases. They're modern day, but the intent is to test drive how these models get applied each and every day.
Gibson Biddle:
So, I'm going to start with these frameworks. I just want to talk about why I focus so much of my life on strategy, and really it helps product leaders to do one of the hardest things, which is to communicate an inspired vision of the future, and then for me innovation and invention is hard because you have to combine chaos, that's needed to invent stuff, there are surprises, but also discipline to make sure that things happen helping you in the way that you want. That they're not random. And I love strategy for its ability to help blend these two forces.
Gibson Biddle:
At the end of the day, for me product strategy is about forming hypotheses for how you'll delight customers in these hard to copy, margin enhancing ways. Margin enhancing is just a fancy way of saying, hey, how to make money. And then of course strategy helps facilitate prioritization. So, as product leaders we can do anything, we just can't do everything. So, imagine the hundred things that you think are important and how do you whittle it down with the top five or ten, and that's where strategy is super helpful. And the last thing, it helps to communicate a plan. So, I'll be sharing how I do these things.
Gibson Biddle:
So I just want to give you a glimpse of three different frameworks. The first, I call it the GLEe, the second the GEM, and then you've already heard me talk about the DHM a little bit.
Gibson Biddle:
But this GLEe model, it helps provide a long-term vision, and it's really about helping companies to get big in their first 5 or 10 years, and then lead a second chapter of the life, and then in the third, this is over, it may be a 15 or 20 year period, to expand even further.
Gibson Biddle:
The GEM model helps different functions like marketing or finance or product to get aligned with each other. And I'll ask a simple question. How do you prioritize these three forces? Growth, engagement, and engagement is really a proxy for product quality, and then monetization.
Gibson Biddle:
And then the DHM model. That's all about delighting customers in hard to copy margin enhancing ways.
Gibson Biddle:
And the purpose of each of these models. The first is to get people thinking long-term, because in the longterm anything is possible. The second is about building cross-functional alignment. And then the third is helping folks to form hypotheses about how to compete in the longterm. So, I'm just lightly introducing these three models, because in the second chapter I'll share how they get put to use in the context of Netflix's current product strategy.
Gibson Biddle:
But I want to dive deeper into that first model, the idea that your job as a product leader is to delight customers in hard to copy margin enhancing ways. I just want to share a little bit more about this, and I'm assuming most of you are subscribers of Netflix, Thank you. So, I'll a list of things that over time we discovered that really are truly delightful for customers. And this is where, I'm hoping that the things that you enjoy about Netflix today are on that list. And then the second is the ability to help build a business, to create profits so that you can reinvest in making an even better product in the future, and that's what margin enhancement is for me.
Gibson Biddle:
At the beginning the Netflix model was actually a la carte. You'd spend four or five bucks for a DVD that would show up in the mail five or six days late. I mean, my theory is all startups suck in the beginning, but the company almost out of desperation discovered this all you can eat subscription model, and that's still what it is today. Lots of price and plan testing. You've probably noticed that over the last four or five years, your price of your service is going up a dollar about once a year, and that's really about helping generate more profits, so the company can invest even more in original content.
Gibson Biddle:
But one thing that's not as apparent is another tactic that's really important for the company today and I called this right-sizing original content investment. So Netflix, they want to have all stories be told on their platform, but they just want to make sure they pay the right amount. And because they have the member tastes of 183 million people worldwide, they're pretty good at guessing, or forecasting how many folks will watch something, and then lining up the cost of investment with it. So, for instance, I'm assuming that almost all of you have watched Stranger Things, and Netflix guessed that about a hundred million people would watch Stranger Things, and because of that they were willing to invest 500 million bucks in the series, and so they right sized it.
Gibson Biddle:
Then I'll take a more arcane title, and now I'm revealing my quirky movie tastes, BoJack Horseman. The guess there was that 20 million people would watch it, and the right sizing meant that they could afford to invest a hundred million bucks. So, just think about the huge advantage that this is for Netflix. They want to provide everything, but they just do it at the right cost, and no other company on the face of the planet has all that personalization data that powers that forecasting.
Gibson Biddle:
And then the last question here is, what makes Netflix hard to copy? So, I want you to think about that for a moment. Imagine I give you 500 million bucks to create a startup to compete. What would you find hard? And for me there's a few key ideas that I'm guessing you're all thinking about, original content, or maybe some of you jump to the brand, but there's really four main buckets.
Gibson Biddle:
So, the first is the brand. You trust Netflix with your credit card, and for years it's been building a product that describes this brand promise, that's movie enjoyment made easy. There's lots of unique technology, but I've already highlighted one piece of key technology, which is personalization.
Gibson Biddle:
And then Netflix does have a network effect, and what it is is this device ecosystem recognized that every device on the face of the earth that's got a screen is magically pre-wired to let you watch Netflix. That's that's the anytime anywhere component.
Gibson Biddle:
And then yes, I'm guessing that many of you were thinking about original content, and that's correct, but the real hard to copy advantage is this huge economy of scale. So, Netflix by virtue of having more than 180 million will be spending about 20 billion in content this year, where poor Amazon Prime is only going to spend 8 billion, and Disney+ 5 billion, and poor Apple only 2 billion. But that's the advantage of being big and that's an economy of scale.
Gibson Biddle:
So, I've shared the model. So think for a moment your job is to create these hypotheses about how you will delight customers in hard to copy margin enhancing ways. So on the left, there's 15 or 20 different theory strategies, then over the last really 20 years. And then I've done a column for delight, a column for hard to copy, and that right-hand column is margin, and I've just checked which one worked or not. So the green is good. So you can see those first five strategies that work in large part as they did, the troika, those three things.
Gibson Biddle:
Then the yellow stuff, you'll see that there were some things. We actually tested advertising, it worked. We were selling used discs. This is in the yellow section. You can see that the check mark there.
Gibson Biddle:
And then the lower, the red stuff. We actually experimented with social. The idea was you'd connect with your friends, we'd build a hard to copy network effect. That would be good. It just didn't work. So all those red things were failed hypotheses. And notice that at the end of the day Netflix got it right maybe half the time. And this is hard and half the time is good enough.
Gibson Biddle:
And at the very bottom, there's some current experiments. Netflix is experimenting with interactive stories with AR, VR. It's not known yet whether that'll work or not.
Gibson Biddle:
So, now I'm going to move on to the second chapter and I'm going to apply some of these tools, frameworks, and models to Netflix. And I'm going to pretend, I'm going to go in and out of character here, that I'm the VP of product today. So imagine it's January of 2020 and I stand up at an all hands meeting and I say, "Hello. I'm here to talk about product." Of course, this is the positioning of the product, and for years we've been working to build a product that, that delivers against this positioning. And then the brand promise, the idea that my marketing partners wanted to tattoo to my forearm, is this one simple idea. Movie enjoyment made easy.
Gibson Biddle:
And then the product vision over, gosh, 23 years, first step, get big on DVD, that's the G in GLEe. The second is lead streaming, and now you have a digital service, and now because it's digital, you can expand worldwide. It made no sense to expand the DVD by mail system. And then the fourth chapter, I think most of you are aware, and this is a strong current focus today, is all about original content.
Gibson Biddle:
So now I want you to think about what's the next chapter for the Netflix product 5, 10, 15 years out. And you can see a little gleam in the eye of Netflix today. One of the theories is that they will be competing in essentially interactive stories, that over a millennia storytelling has changed with technology, and this'll be the next step. And you can see some signs of that. I'm guessing a few of you have watched Black Mirror's Bandersnatch. That's an interactive film. Or, this month Netflix launched the Unbreakable Kimmy Schmidt, which is an interactive film where you can help Kimmy decide if she could just make out or plan the wedding.
Gibson Biddle:
And then the high level metric that all product leaders in Netflix are focused on is retention. And this is the high level engagement metric. And for perspective, at the startup phase, the company thought 10% of its customers would cancel. In the early days in 2005, about 5% of the customers would cancel each month and today it's about 2%. So, retention has gotten dramatically better, and the idea is through these combinations of delighting customers in hard to copy margin enhancing ways, it can get even better. And every 10th of a point at this point is huge for making the business better.
Gibson Biddle:
So in my first month back, I would arm wrestle with my marketing partners, with technology partners, with finance, asking a very simple question. How would you prioritize these three forces? Force rank, growth, engagement, engagement is a proxy for product quality, and monetization. And after a lot of debate, we settled on the idea that growth should be first, and the goal is year over year member growth of 20%. Monetization, and that's essentially economic efficiency. The lifetime value for a customer today is 170 bucks. Perspective, it was a hundred bucks in 2005. Keep moving it up to the right, largely through improvements in retention and some price increases. And the third is engagement. Make the product better. And the thinking is that 2% monthly cancel rate is very low. So, this is the hardest thing and there might be some diminishing returns.
Gibson Biddle:
And then the current high level strategies that the team is focused on today There's probably six or seven, but I'll highlight the key four. I've talked a little bit about an importance of personalization, about original content, a little bit about interactive storytelling, but I want you to key in on the watching experience. And so for each of these high levels theories, for these strategies, there's also a proxy metric. So, the proxy metric for watching experience is the percentage of customers who watch at least 40 hours in a month. Retention's a great metric, it's just super hard to move and it's hard to do big A/B tests to really understand what moves the needle or not. So you use these proxies, these lower level metrics that are easier to move, to try to get some insight on whether the strategies are working or not.
Gibson Biddle:
And then the tactics or projects against watching experience. Ultra HD, even better video quality, sound quality, custom playback speed. I'm going to talk about a case of shared viewing in a moment, essentially Netflix party. Or some of you may have noticed that they're getting better and better at their lip sync, taking original content in one language and translating it into 40 other languages. And I call this the SMT lockup. The strategy on the left and metric proxy, and then the tactics or projects.
Gibson Biddle:
And then this is a rolling 2020 roadmap, Q2, Q3, Q4, and that last quarter's Q1. And you can see how the projects will play out over time against each of those strategies. In a moment, I'll talk about the shared viewing project for Q2 against the watching experience lineup. So this is a plan, and this is a way to communicate the product team's focus. And with that, I'll say, "thank you," and then I'll switch back to Gib. That's me.
Gibson Biddle:
So now this third chapter, I want to share two Netflix cases, and I want you to put this in the position of, "Okay. If I had made the decision, what would I do?" But I really want you to see how these frameworks can help your day to day decision-making.
Gibson Biddle:
So the first case, I call it Netflix Party, and amiss COVID, there's lots of folks that are isolated, and there's actually a Chrome extension from a little team that's working under the guise of Patreon and it's called Netflix Party. You can see it here. It lets Netflix members all over the world watch movies together, and you can see this is Stranger Things. So you can see on the right there, they're chatting, they're giving each other grief, they're clapping, whatever. So they have this social experience where people all over the world are watching the same movie at the same time.
Gibson Biddle:
And the question I'm going to put to you today is, should Netflix execute this idea? How does it work? For this little Chrome extension, you install it. That won't be necessary for Netflix. You open up a video, you create your party, and then you make it easy for other people to join and watch the same movie at the same time. You know, if I nicely asked most folks this point, I'd three quarters, maybe four fifths will lean into it. Like, "huh, this seems like a good idea, especially during isolated times," but I just want to give you a little bit of consumer insight, largely about the Netflix social experiments over the years. And it's a dirty, messy history.
Gibson Biddle:
So, here are all the things that Netflix killed. So we did experiment with a friends feature, and then in 2009, we killed it. Basically we couldn't get more than 5% of our members to engage in the feature. And to my thinking, if you're going to have a shot at improving retention, you need to see stuff that gets used at least 20% of the time, and it was only a 5 percenter. And then same thing happened. There was actually something called Party Mode on the Xbox and Netflix killed it in 2010 as well. Again, less than 5% of the members engaged.
Gibson Biddle:
And then this is a little different, but as Netflix was expanding internationally, lots of different countries, they didn't have big budgets. So they experimented with a feature, instrumented word of mouth, essentially. If you were in France and you saw movie you loved on Netflix, you could suggest it to a friend, and if they were on Netflix, that was easy, but also it made it super easy to disseminate, be it a text or WhatsApp or Twitter or Facebook or whatever, it's just magical integration. And they killed it. And my guess is, again, it wasn't a well used product.
Gibson Biddle:
So what's up with social at Netflix? The main insight is twofold. First, your friends have sucky movie tastes, and I think intuitively, you know this, and then there's a second thing, which is we all have lots of guilty pleasures and we don't really want everybody to know everything that we're watching. So this is the cold water about these social ideas. So here's the decision today. Should Netflix execute this idea, and that the team by virtue of these well-crafted application programming interfaces is able to execute it. It's actually quite hard to copy. It would probably take years before Hulu or Disney+ or Amazon Prime could execute this.
Gibson Biddle:
And there's actually some pretty amazing data from that Patreon team. So in the last 60 days, the number of folks that are using that Chrome extension, and you can imagine it's hard to find plugin et cetera. It's gone from 500,000 to a million. And then if you look at the download history in the Google Play store, there's been 10 million downloads. So that's a sign that this thing is going up into the right. This is a substantial proof of concept. And so the team can launch Party on laptop next month, and then they can extend it to all platforms by the end of the year, and they're ready to do it.
Gibson Biddle:
The question I want you to think about is, will this delight in hard to copy margin enhancing ways? The way to think about this is will lots of folks love it. I've already told you that it's hard to copy and the margin is going to come from two tactics. One, people would rave about this and build word of mouth and encourage even more folks to join around the world. Or, the second, this could get used by enough folks that it's got a meaningful shot at improving retention and taking it from a 2% cancel each month, to bringing it down to only 1.9% cancel each month, and that is incredibly meaningful for the business. That would build margin if that could happen. So the other way of thinking about this is, should Netflix try social again? You could see their willingness. They've experimented multiple times. Keep trying. Should this be the next experiment?
Gibson Biddle:
So the question I'll ask of you is, if you were the product leader at Netflix, would you launch this or not launch this? And that's the question I have for you. And I'm able to do this with large gangs, and if I put it to folks 63% say, "yes, let's do it," and 37% say, "no." So, I'm going to reveal what I would do. The key thing, there's no right or wrong answer on this, but the key thing is the thinking behind it.
Gibson Biddle:
So I would choose not to launch, and I want to just give you my thinking. The first is, yeah, two thirds of us are into an idea like this, but we are not normal people. Right now, you're participating in an online conference. I built this slides using Google Slides and recording in Zoom. I did that polling using Slider. We are not normal people. And this is the reason that whenever we would do focus groups or qualitative or usability at Netflix, I would never do it within Silicon Valley. I would go to Memphis, Tennessee or Providence, Rhode Island. So, I'm just guessing we overvalue this thing compared to normal people.
Gibson Biddle:
And then to create a simple service, you've got to say no a lot. The reason that Netflix killed all those things, we call it scraping the barnacles, and the barnacles are on the bottom of a rowboat and they just slow you down. So, we were disciplined about getting rid of all that stuff in order to create that simple service. And in this case, my guess is it would be another five percenter, and without substantial delight there's no real opportunity to build hard to copy advantage or margin. And I'm guessing if you launched it, you'd see it's yet another feature where it only got 5%.
Gibson Biddle:
And the last idea is if your job is to build this experience where movie enjoyment made easy, none of this is actually easy. So think about this. Think about how hard it is to pick a movie that everybody wants to watch at the same time. Netflix lets you stream three or four things concurrently in the house, because they know how hard that is. You know, you've got mom and dad watching up in the bedroom and one kid in her bedroom and the other kid, this is my family, my 8, 22 and 24 year olds watching in another room. It's just really hard, and so these are the reasons I'm not supportive of the idea.
Gibson Biddle:
So I have a second case and this is pretty simple. The result's known. I'll tell you what happened. But this idea is, should you auto cancel inactive members? And the thinking here is you're a bright product manager, you look at the data and you notice that about one half percent of all of Netflix members haven't actually engaged in the service in the last 12 months, but every month you're hitting their credit card for 13, 14, 15 bucks. And the question is, should you do this? Or should you just reach out and say, "Hey, we're canceling your service."
Gibson Biddle:
So this is what the bright product manager was thinking about. And the assessment was, okay, let's send an email, a notification to these folks and ask the simple question, would you like to cancel? And if they say, "no," we won't cancel them. But if we don't hear from them, let's just assume that they don't really want the service. They signed up. They forgot about it. So, that's the idea. And the impact of this, if you were to do this, if you auto cancel all the people that are hanging around and not enjoying the service, not using it, you're going to lose a hundred million bucks for the company in the course of the year.
Gibson Biddle:
So, here's the question. Should you do this? Will this delight in hard to copy margin enhancing ways? And we already know the margin story, which is we're going to lose a hundred million bucks. Now, let's think about the delight. Maybe this is the right thing to do. If you learn that Netflix chose to do this, you'd feel good about them. And then the other question is, what's the hard to copy advantage at work here? I'll let you think about that. And I put this question to folks all the time. So if I ask again, whether you should launch or not launch, their assessment is maybe you should? And I really want you to thinking about that question yourself. You can see in this last 300 people I asked, 53% said, "yes," and 47% said, "no."
Gibson Biddle:
So I'll tell you what I would do. I would launch this. And my thinking is as follows. The hard to copy advantage that you're building is the brand, and guess what? This is a real case. Netflix executed this about two or three weeks ago, and the assessment was that it was worth spending those hundred million bucks. And it's not like you're really spending them. Maybe Netflix shouldn't have been getting that to begin with. And the reason is this is the cost of doing the right thing, of doing the ethical thing, and of building an even stronger world class brand that builds more word of mouth and builds more hard to copy advantage.
Gibson Biddle:
The other thing that's interesting is for most folks, a hundred million bucks sounds like, if you were the product manager making this decision, it sounds like a ton of money, but the project manager actually said, "No, it's not. This is a low stakes decision." It's only a hundred million bucks against a company that's going to do 20 billion in revenues this year, and by the way, it's totally reversible. By definition, low magnitude, and it's a reversible decision. We can undo it. It's a low stakes decision and they just did what Netflix does. It has a quick debate, but then decide and do, and it's happened. You can see it happened a few weeks ago.
Gibson Biddle:
So I'd like to bring things home for you with just a couple of ideas. So I've been talking about strategy, and strategy is about this exploration, it's about this experimentation, to find a route from here to there, from a punk kid DVD by mail company, to what I think is a world dominating streaming service. You can see this woman beginning to explore and go to the ocean.
Gibson Biddle:
And so, at the end of the day, these strategies are about forming these hypotheses about how you'll delight customers in hard to copy margin enhancing ways. And then I shared three frameworks and went and tried to lightly expose them to you, but they let you combine the discipline, they bring discipline, but not so much that you squeezed the life out of the chaos and the invention and the innovation that's required. And then all of these tools enable you to communicate an inspired vision of the future, and that's probably one of the hardest things there is to do as a product leader.
Gibson Biddle:
So I'm just going to bring you home with one last idea. If you want to build a ship, don't drum up the people to gather wood, divide the work and give orders. Teach them to yearn for the vast and endless sea. And that's what well-executed strategy can do. And this is from Le Petite Prince.
Gibson Biddle:
And so with that, I'd like to say, thank you. Thank you Product-Led Summit for having me. Obviously a Stranger Things theme. If you know me, I'm not actually quite done. There's one more thing that I'd like to do. Sometimes, I feel like a street performer. I'm doing my best to breathe fire. You've all been in the situation. You're in the crowd and then the street performer passes the hat, and this is that moment. You know, I usually skulk away, but in this case, I'm not asking for money. I would love, love, love, love your feedback.
Gibson Biddle:
So, you can hold up your phone right now and this QR code will activate a SurveyMonkey survey. You can click on the link, and then it's asking a simple question on a scale of zero to 10, where zero sucks and ten is great. How likely would you be to recommend this talk to a friend or colleague?
Gibson Biddle:
And then there's one other question, which is, "Hey, what was good about this talk", one thing. And then one idea that would make it better. So, this is my 534th SurveyMonkey survey. This is my proxy metric for talks, and I would love, love, love your feedback because it's incredibly helpful to me. If you have an Android device with Lollipop that didn't work, you can just go to GibsonBiddle.com and I've provided a link there.
Gibson Biddle:
I've also provided a PDF of the presentation for you. And then I wrote a series on Medium, How To Define Your Product Strategy, where I go into depth on 12 different frameworks and models. Nothing bad will happen. I don't collect emails, I don't do advertising, so that stuff is just there to be helpful.
Gibson Biddle:
So, with that, I truly say thank you. And as I leave, I reveal my own guilty pleasure. This is a current hit, Tiger King on Netflix. So I'm revealing, yet again, my bizarre taste. Thanks so much Product-Led Summit for having me, and I hope you found this helpful. Goodbye.
