 Hi, everyone, excited to be here today and thanks for your time attending today's webinar. I am Andrea Wang, currently a senior growth PM at Amplitude. And today excited to chat about how to drive non-linear exponential product growth. And before we get started, just want to briefly introduce myself. I started my education at Stanford after obtaining two degrees. I was a management consultant at Kearney, where I mainly worked on advising Fortune 500 companies in fintech and consumer verticals on their product innovation strategy. And then I joined Lyme pretty early on after series A, wore multiple different hats in the company's early phases, was working across government relations, strategy and marketing, and later on transitioned to become a full-time product manager on the core rider product team responsible for international expansion on the rider team. And from there, I decided to join Amplitude because I was such a power user of the product as a Lyme PM, and I was really blown away by the type of impact this product was able to provide me with as a product manager. So joined Amplitude to become a founding member of the growth team and have really enjoyed scaling with the team and scaling the team's charter within Amplitude as well to grow it into a very prominent pillar within Amplitude. So today, I'm here to talk about product growth, definitely one of my favorite topics. So we'll start by talking about the concept of product like growth, what it is and what it isn't. And then we'll discuss two common frameworks that I often hear teams reference when they talk about product growth and discuss the trade-offs. And finally, given I am a practitioner of product growth for many years, I want to give all of you some actionable tips and frameworks that you can come away with to refine your current practice for product growth at your company, but also in case you're starting out from scratch, hopefully these are some actionable tips for you to get started as well. So let's dive in. Product like growth is definitely a very popular term or concept that I keep hearing about out there. But the issue is every person I ask, I usually get very different answers in terms of what it actually is. So first, let's talk about what it isn't. First product led growth does not mean being product manager led. Great products really emerge from empowered fully cross-functional teams, and there's really no perfect one fits all leadership model. So it's not that. Product like growth also does not imply some sort of functional hierarchy between product and other functions within a company, such as marketing and sales. And in fact, you need a lot of cross-functional alignment and support to make a product like growth motion work for you, even for products without any sales touch. And then finally, product like growth is not about completely removing sales or marketing in a picture. In fact, in many product like products, marketing and sales play a really critical role in helping users get value and close to sale. And typically in very product led companies, you also have very differentiated go to market strategies for certain segments of the customers. For example, sometimes for your enterprise customer base, you actually need a lot of hands on nurturing and sales touch in order to close the deal or expand them. So now that we have talked about what product like growth isn't, let's talk about one of my favorite definitions of what product like growth is. This is from a preeminent source on PLG OpenView partners. And they state that product like growth is an end user focused growth model that relies on the product itself as the primary driver of customer acquisition, conversion and expansion. And what OpenView means by that is in the past, there's been a lot of focus on marketing, for example, as the key acquisition channel and identifying what channels are bringing people into the platform. And those things are still important and don't go away. What OpenView's definition of product like growth suggests is that there are a lot more levers that we can pull within our products. And by levers, I'm referring to when someone finds enough value in your product that can drive them to increase engagement, the users do the hard work of acquiring other users for your product because they are getting so much value. And that's what product like growth ultimately is trying to drive. And today, I'll try to share some actionable frameworks and tips that I think apply to a wide range of business models and products that hopefully could be actionable for you all. And now let's talk about two of the common frameworks that people usually reference when they think about growth for your products, a growth loop or a funnel. And let's talk about what we mean by that. This is a typical way to think about a customer acquisition funnel. You have marketing in an organization usually being accountable for acquisition part of that journey. And their KPI is generating leads or new users. And then from there, many organizations view product as a function, owning, activating those users and also retaining them. And typically, product is accountable for KPIs related to product usage and engagement. And for example, active users could be a common KPI that product is held a combo for. And after that, after retaining a customer, you have retention, which is owned by sales. So sales is generally being accountable to generating that initial sale, but also retaining and expanding those customers later on as well. And in a lot of organizations, this is how these organizations view the success for each function. And there is usually a handoff point between different stages in this funnel. Now let's talk about an alternative framework to think about product growth. Growth loop is a concept pioneered by the experts at Reforge, a professional development program for tech professionals. And the key characteristic of a growth loop is that you would have certain input actions performed by usually new users in your product that lead them to perform certain actions, which eventually leads to certain outcomes. But that output also feeds back right into the input and gets reinvested so that this results in more and more users performing the input action. And how this framework of thinking manifests in an organization is that teams need to collaborate across functions to really make this type of growth loop working. And instead of, for example, marketing being solely accountable for acquisition and product being accountable for activation and retention part of the journey, you would have cross-functional tribe of marketers, product managers, engineers, designers within the pod, for example, working on acquisition together. And then jointly they would figure out what are some product-led acquisition growth loops, what are some marketing-led acquisition loops, and what are some sales-led acquisition loops and have these cross-functional team members being jointly accountable for certain KPIs that are shared across functions. So as you can tell, this is a very drastic contrast with the previous mental model of thinking in funnels. And it really manifests itself in terms of how the org is designed and how teams collaborate with each other on a day-to-day basis. So the question then becomes which one is better? So I would like to pause for just a few seconds for everyone to think about this question. Should you think about your growth model in funnels or in loops? We'll pause for a second and then we'll dive into the one that I personally would highly recommend everyone adopt in terms of thinking about product growth. All right, hopefully everyone has an answer in your minds and time for the reveal. I strongly believe that every growth team or product team in general should think in loops, not funnels. And the reason why I personally think funnels is not an ideal framework to generate compounding sustainable growth for your product is that when you think about a funnel, it has a beginning and it has an end. So typically in an organization, the way you solve for how do we grow revenue is let's get more marketing dollars to invest in more campaigns and hire more sales reps so that we can go after more logos. But in that framework of thinking, the funnel makes it such that the more you invest at the top, it results in a very finite set of outcome at the end, which is ultimately the number of customers that you are able to acquire, retain, and monetize. And this also results in a lot of investment into channels or tactics that are not scalable for your company in the long run. And for a team that's thinking about growth, it's really important to prioritize areas where you are able to generate sustainable, long-term, and defensible growth for your company. And that's why the growth loop framework is a lot better in that way. You're trying to figure out what are some loops that may exist in our business that is self-sustaining in nature, that you don't have to constantly pour in more headcount, in more add dollars, and more budget somewhere in the organization. But instead, you can figure out what are some ways we can sustain growth long-term by making sure that your product is built in a way that can help you grow more sustainably. So let's actually put it into action to look at what growth loops actually look like. Let's take Slack as an example, because I'm sure many of you are really familiar with the product. The key action that they want a new user signing up for Slack is to create a Slack workspace, to really start on the journey of adopting Slack's core features. And this may become really clear to the user signing up that Slack will be 10 times more useful when used alongside other colleagues. So the output that they want the user to take is then to invite other colleagues to the same Slack workspace so that they can start collaborating together and communicating together. And the colleague being invited sees the invite via email and accepts the invite, which makes him or her a new user, which then starts the growth loop all over again. And once Slack gets this loop working, as you can imagine, they don't need to pour a lot of marketing dollars and hiring sales reps just to make sure that Slack is being adopted by more and more users within an account, but also growing the market share, because once users start signing up and getting value from one workspace, you can imagine that as they go on to other companies in other roles, and also start using Slack across more and more use cases beyond professional use case for just their day job, but also across maybe like social work, social use cases, Slack will be adopted to buy more and more companies, simply by users getting value and adopting these products for the very first time. Let's take another example. And what better example to use than our own amplitude? So we're talking about our analytics products, since amplitude has really expanded our product portfolio to be a multi-product company now. But for the analytics product, when a new user signs up for amplitude, they typically want to measure user behavior of how users are using one of their digital products. And after signing up, we want to get them to an insight about their users as quickly as possible. And there are a few ways that our users can get there. For example, they could view an existing chart that is made by another team member, or they could learn the taxonomy and create a new chart of their own from scratch. But ultimately, we want them to share a piece of insight with their colleagues. And they typically get that by looking at a piece of insight that they previously maybe didn't know before about how users are using their product. And once that insight is shared, this has an acquisition impact for amplitude, because more and more users within the same company are exposed to amplitude insights. And those invited users will in turn become acquired new users onto the platform. And this cycle begins again. And once they get value themselves in counter pieces of insight, they would then help us distribute those insights with other people within the organization. And sometimes there are very powerful growth loops that are kind of interconnected with each other, and they have impact beyond just one part of the growth model. For example, using again amplitude as an example, we believe that the more users that are creating and consuming insights and amplitude, the more likely they will encounter interesting insights that they want to share with their teammates, what we talked about just now. But that also means that as more of the team exposed to amplitude insights, they grow to know the value of amplitude and become amplitude users themselves. On the other hand, if we have more and more users within the same company actively leveraging amplitude for insights, the more they will need to instrument their product so that they can get more granular insights about how users are using their product, the more features, especially the more advanced features they would need to use to answer some of their questions about user behavior. And that really enhances our ability to monetize our customers. So in this way, our customers value out of amplitude scales really well with the value that amplitude receives from our customers in return. And typically, when you encounter scenarios where you find certain growth loops that are interconnected with each other within your business model, those tend to be the ones that are super important and high priority for your team to prioritize investing in and making sure that they're working really well. Now we've talked about generally what is the framework that I recommend everyone to use in thinking about driving compounding sustainable growth for your products. I would like to share some of my tips on how to actually put this in action and start to actually build for growth loops for your products. Step one of this process typically is thinking about qualitatively what types of growth loops, if any, already exist today. For example, you may already have an invitation flow in your product where existing users can invite others that are not part of the product already. So you may already have a viral growth loop for your product. Or you may have certain sharing features within your product that enables content to be shared with non-users externally that could have an acquisition impact for your product as well. At this stage, I would encourage everyone to qualitatively map out as many of these growth loops as possible, even if you're not sure if they're actually loops and functioning like that today. And also feel free to start brainstorming what new loops you could build for and design for in the future, even if they don't already exist in your product. Maybe you got some inspiration from maybe the presentation today or case studies that you have read up elsewhere from products in similar industries that you feel like are missing from your product. Try to document those as well, because it's always possible to consciously design and build for those loops in the future. And then it's time to prioritize which loops are most important for your product so that you can invest more time in quantifying them, understanding how well they're working, and where to invest your resources for the highest impact on ROI. Let's say that you believe that viral growth loop is one of the most important loops for your product due to the low cost of acquiring new users and the potential impact to your growth if this works really well. And maybe because you believe that your product has a very strong network effect and that users can get more value if more users are on the platform, which also leads to better monetization at the end of the day. You need to first audit the step-by-step user flows with the current invite experience looks like, and what are the potential ways someone could drop off in that process and make sure you instrument every step of that loop so that you can get very granular insight into how users are converting between the steps. For example, make sure you're tracking how many new users are being acquired in each time period. What percentage of them send an invite during that time period? And on that average, how many users do they invite per person? What percentage of those invites get delivered via multiple channels? What percentage of them open the email and end up clicking through? And what percentage of them ultimately signs up for the product and become an active user? And let's actually put this into action and look at a hypothetical social app as an example. So when you're starting out quantifying one of your growth loops, one of the first things you need to look into and decide is what time period is the right one to measure a cycle of growth loop for your product. This typically maps to the usage frequency of your product, which is how frequently your average user is actively using your product today. For some products, it could be daily, for a lot of social network products. For some products, it could be weekly, for some productivity apps. For some products, it could be longer, like monthly or yearly. So I think knowing your average usage frequency is really helpful. But then I would also recommend you look at the time to distribution, the time to convert from being a new user to performing the beginning input action. In this case, let's say, inviting a user, because you also want to understand maybe your users are using your product on a weekly basis, but does it typically take a lot longer for a new user to get to a new user? For a new user to get to the point of wanting to invite someone else? Because maybe your product is really complex to understand. So on average, you actually might see new users take way longer than the first week after sign up to get to a point of sending out the first invite organically today. So after determining the cycle of, which is the time frame that you should be using to measure different steps in your growth loop, you can then pick a starting time period to start modeling out the loop step by step. So let's look at this hypothetical product as an example. In month one, you may have acquired 100 new users. You see that 10% of them send an invite. On average, each user invites two new users, and 40% of those being invited were delivered the invitation via email, opened it, clicked through, and completed the sign up process to become a new user in your product. This means that at the end of the first month, you have eight new users as the output of the cycle being added to your product from existing users invite. And those users are subject to another cycle, and a certain percentage of them might initiate that invite flow again, which in turn, acquires another set of new users. Unfortunately, when you look at this example, this loop kind of dies after month two, because there are not enough new users being generated at the end of that second month to allow the loop to spin another cycle. In this sense, this loop isn't working that well today. And as much as you would imagine this being a growth loop because you think, you have an invite feature, and there is a sense that there must be a lot of new users being acquired that way, in reality, it's more functioning like a funnel today that has a beginning and has an end. Whereas a true growth loop that is working well should see the output feedback into the input and have at least a couple of self-sustaining cycles that can help generate sustainable growth with every metric that you use to measure the impact of the growth loop. So this one needs a lot of work. But after modeling this out, what this does is it gives you a very clear sense of how well certain growth loops are performing. And this way, after doing this for more than just one, you can then compare several growth loops together to decide which ones are working well, which ones are not, and which ones are worth your team's time and investment to make sure that you improve their performing. So let's just take the example that we just saw before. It's very clear from the metrics that we modeled out. There are two top constraints in that particular buyer growth loop. Number one was the percentage of users sending invites. The vast majority of your active users in that model weren't actively inviting other new users. And the second top constraint in that particular growth loop was percentage of invites being accepted. Maybe you have done some benchmarking against other products in your category and learned that the percentage of invites being accepted is really low in terms of what you're seeing today. And after doing this, you then reached a realization that we need to improve these particular steps in this loop because if we are able to get this to perform, let's say 50% better than what it was from before, this actually has the potential to perform a real growth loop versus a funnel today. So you might host a brainstorming session with your team and come up with some ways you could drive more users to send invites and drive up the acceptance rate of those invites. For example, you may realize that your product does not have a guided onboarding flow that nudges a new user to invite their friends after they get some initial value from your product and reaches their aha moment. So your team might want to run an experiment on that to see how it drives variety and engagement. And also after rolling down into why invites are not being accepted step by step, you may find that you currently have an email deliverability issue with those invite emails going to spam. Or also you might find that the click-through rate of those invite emails is really low, which means the invite email itself needs a lot of work. And it doesn't do a great job of conveying what this product is, why someone is being invited, and what value they could be getting by accepting the invite. And then you can really focus on specific tactics and experiments to try to improve your hypotheses and improve those specific parts of the funnel that you identified to be the highest points of leverage to improve this growth loop in general. And after that, it's just a matter of constantly iterating and experimenting and measuring how this loop gradually improves over time. And you may decide at some point after running a few experiments that this is not a meaningful lever for your product anymore, or you have gotten it to perform well enough, beyond which you kind of hit some diminishing returns. And it's no longer worth your team's investment investing further. But that is the general process that I would recommend you all to take, especially as you're starting out to think about measuring and understanding a lot of your growth loops in your product. Or when you're in the optimization phase, you're trying to improve on how those loops are working. This process also applies as well. So we have covered a ton in this webinar in such a short amount of time. So just want to recap on some key takeaways that hopefully everyone can take away from this. First of all, if you were to remember one thing from this, the main thing is that thinking in loops is better than funnels. Thinking in a funnel approach tends to result in non-scalable activities and tactics, but growth loops really force you to think about how to create compounding sustainable and long-term growth for your products. And in terms of how do you build for growth and build for those growth loops in your product, I would recommend everyone map out what loops might already exist, first in a qualitative way, and then try to quantify some of your loops one by one so that you can really compare how well they're working, understand their current impact on your growth, and also identify key points of leverage that you and your team can spend your time and resources on to get the highest amount of impact for your business. And even if certain loops are not currently present in your product, that's totally fine. It's possible to design your product differently to create them and build for them consciously going forward as well. So that's it for today's webinar on driving nonlinear exponential product growth. Hopefully the content is helpful and actionable for many of you and definitely open to further connecting and discussing some of the content and exchanging insights around product growth. So feel free to add me on LinkedIn or shoot me an email if you want to chat more. Thanks everyone, and thanks for your time today.