Making your Product Strategy Accessible with the North Star Framework

0:06 Hello, my name is John Cutler, and I am extremely excited to be presenting to you today on the North Star Framework and how we can use it to make strategy more accessible. Very brief introduction about me. I work at a company called AMPLITUDE. At AMPLITUDE, we help teams do more impactful work. We use product analytics and experimentation and recommendation, etc. And my role day to day is helping our customers and future customers be more effective both with our product, but also just more effective product developers and product makers and product managers. So it's a great role. I get to interact with lots of teams on a day-to-day basis, and I get a good perspective about how teams are working around the world. So, our topic today is making strategy accessible with the North Star Framework.

1:06 I hope you enjoy it. Very brief goals for the talk. By the end of the talk, hopefully you'll have a better idea of what the North Star Framework is and then how you might be able to start using it at your company to let the people who are working on your products, executives, really the whole company be better aligned on what you're doing, why you're doing it, and how you can have impact. So let's get started. So I always like to start with the problem. Before we get into frameworks, what's the problem? Now there's many different frameworks that actually try to solve this problem. North Star Framework is particularly good at it, but let's zero in on what we're talking about. So if you imagine a spectrum of work, so imagine we've got at the highest level, some amount of business strategy related to the company. And then on the other side, completely, you have the work, you know, the experiments, the things people are trying, the things people are building, the things people are designing.

2:09 And now if you think about one level higher from that, you might have what we would call the opportunities or the problem statements or the problems we're trying to solve. And if you think about this work together, the business on one side and then the work on the other, often we have what I would call a messy middle problem. And it really is around here related to the product strategy. And the problem here is that, you know, at the highest level, the business has slides and things like that that say what our goal is. And then our roadmaps and JIRA and all kinds of things have tickets and stories and epics. But often the product strategy itself is in presentations. It might be very high level, you know, go up market or meet the needs. of this customer. And it's really difficult to connect our work and connect the problems we're solving to that business strategy. So we want to connect the business strategy to what we're working on on product.

3:15 And often the product strategy itself doesn't do this. And it leaves what I would call kind of a messy middle problem. We just don't know what is happening here. And it makes it very difficult. Now, what we use the North Star framework for is to not replace a product strategy. Certainly, product strategy is a very big topic. But to do what I would call create an accessible entry point into the product strategy. What we're doing is we're thinking about an indicator and a series of metrics. You could call this sort of a constellation of metrics. If you think about the star metaphor or the astronomy metaphor, what we're doing is we're thinking about a model. That allows us to create a leading indicator for business success, but then when we're doing the work that we're doing on the day-to-day, allows us to connect into that.

4:20 So that's kind of the goal of the North Star framework. One other really important part of this is that, you know, if you think about the work we're doing on a daily basis, You know, like one to three weeks, right? We're working on stuff for one to three months. One of the challenges at working at this resolution is, and people find this a lot with OKRs, is they get done with a quarter and then they say, oh goodness, you know, what are these other problems we need to solve? So you get a lot of shifting every single quarter. What the North Star framework does is it provides what we call a persistent model. So it actually lasts for years. A stable model that we can plug in our work to. So this is not going anywhere. This doesn't change every quarter. And so that's another nice element of the North Star framework that solves this particular problem. But the problem is, is that we feel like we're going, you know, back and forth. And what the North Star lets us do is it lets us create this stable element here.

5:23 So hopefully that describes the problem. The problem is how do we connect our work and the things we're doing to the things that the business cares about in a way that is holistic, that lets us have a solid product strategy that can grow over the years instead of just chasing every single quarter. Let me give you an example, and then hopefully this makes the concept easier for you. So imagine for a second we have this company. It's called Do It Yourself Company, and they send people kits to build things at home, like, for example, like these planters here. So who's involved? You know, let's think about how they thought about their North Star. Who's involved? Well, you know, parents with kids, they want a project maybe, or people who are curious about do-it-yourself projects, makers, you know, people aspiring to make things at home. So that's their target persona, who they're focusing on. Now, what promise are they making to the people out in the world?

6:28 What's their product promise? So obviously this varies by person. For some people, they want to decorate their house. For other people, especially parents, I'm a parent, especially during the pandemic, you want something away from screens that you can work on with your child or my son and make something with our hands instead of just being in the screen all day. So that's this job replace screen time. And really the job of this product is fun and curiosity. I don't think anyone will say that making these projects at home is the fastest way to be able to make you know, kits like this, but it's certainly the most fun way that you could do with your family. And then finally, we think about the parts of the puzzle. So for this particular company, which does exist, I've changed the names of this company, you know, what's involved with a great product? Well, people need to find projects that matter for them.

7:28 They need to be able to get support. They need to be able to access a community. They don't want projects which are too difficult or too easy. They want just that right amount of challenge in what they're doing. The kits need to arrive on time. The kits need to be usable, right? You can't, you know, if you can't finish the project, you're in trouble. So these are the types of things, you know, when we think about what are the parts of the puzzle for this company, what are the, you know, what things fit together. So they use this as the basis for thinking about their North Star and inputs. And let's just run through with what they found. So, um, The way that they thought about this is they answered this question, right? Our path to medium to long-term sustainable growth is a function of our ability to be able to do these things. And what they realized at the highest level, they really needed to be able to inspire and equip new, you know, their customers to finish these projects. So that was their...

8:31 Vision for their product and they broke that down into a metric. So we can see here that their metric is repeat makers. So that's the number of people who are finishing multiple particular particular projects, and then they defined that metric as the number of DIY customers who finished, you know, two or more projects in a certain period of time. So that became their North Star. That's what they were aligning the efforts of their company was creating these repeat makers, which were a sign that the project strategy had worked. Now, one metric alone is not enough. And this is one of the interesting puzzles with the North Star framework in the sense that if the metric is too leading, if it's too addressable by the team, then you get a little bit of a short-term effect. And if the metric is too far away, the teams have a difficult time aligning their work around that. So they came up with three things that we call inputs.

9:33 If you imagine the North Star is the output of these inputs, they came up with three parts of the puzzle that were more actionable for teams. So in this case, the first one was project matching success rate. And that was the rate at which that they could match these customers who are arriving at their site or app with the projects that they were going to finish. The second one was project completion rate. And this was the number of people. who actually checked in that the project was completed because they had an app that lets you check in when the project was completed. And then the final part of this was active community members, and that's the number of people who were active in their community at any given time. So you can see here that together, This forms the North Star. So the North Star is repeat makers, it's right here. And then they have these three inputs, which are down here. And those are the three inputs that matter. So you can see from this that what does this help the team do?

10:36 Teams are able to focus on particular parts of the puzzle. You know, you are a team, you come in and you focus on active community members. But then we believe that the more, you know, if we can move these inputs at the same time, this will successfully increase the number of repeat makers. So you get the general idea, you get the general concept of how it works. So what makes a great North Star metric? There's certainly some that are not so good. So I have some examples here. What would a bad North Star metric look like? You know, app installs. It's okay when someone installs an app, but that doesn't mean that your product is working. That's just a very, very, very leading indicator. It's not actionable enough, really. To know that your product is succeeding if you see that number go up. Same thing with revenue. Revenue is kind of curious. Because one of the challenges with revenue is in some ways it's too leading and too lagging. It's too leading in that there's many things you can do from a marketing perspective to increase revenue.

11:43 And also great products often only produce revenue in the future, right? It's working and working and revenue happens in the future. So revenue is sort of too leading and too lagging at the same time, depending on your situation. So good North Stars are things like healthy projects completed or active community members or learning users or the connections between fans and artists. And they have a number of these characteristics. So what makes a great North Star metric? One, it expresses value. So back to our DIY example, we can see why it's valuable because if people are finishing multiple projects, that's a good sign for us. It represents our vision and strategy. Yeah, pretty much, right? It lets us know that if people are... Finishing these projects over and over and using the community, etc., that it's a decent representation of the strategy. It's a leading indicator of success, certainly. If we could demonstrate that people kept repeating their projects, that would be a good signal.

12:44 It's actionable. Yep, it's understandable. We can understand repeat makers. That's good. It's measurable. Certainly, if we go back here, yes, we are able to measure it. So that's decent. It's not a vanity metric. So that would mean, for example, that when this number goes up and down, it means something to our team. It's not something like active users, right, which it goes up. Wow, it doesn't. There's many ways you can make active users go up. So if this number increases as a company, this is going to be a good signal to us. Great. So... Let's think about this a little bit more and why this is an important concept and how it can help make strategy actionable. So if you think about here, we've got our repeat makers and we've got project matching success rate, project completion rate and active community members. That's our North Star and our inputs. What you'll see here is that the team and what they're working on connects to this input.

13:52 So here we have project matching success rate. And an opportunity here could be better understand the user's intent. That's a pretty decent way to improve project matching success rate. And maybe the team comes into work and says, you know, we're going to build this multi-step flow to be able to capture interest and intent. So this might be the work or the experiments they're running on. So what you see here are these levels that we were talking about in the beginning. So, you know, you've got over here the North Star framework. This is the context that will last for quarters or years even. Over here, you've got this opportunity level. So that's, you know, the problem statement or maybe where we think there's an opportunity. And then finally, over here, you've got the intervention level or the experiment level. The important point with all this is that, again, you're linking the work. To the opportunity. There's another important component here.

14:54 So I know a lot of people use things like OKRs or they use other methods. This is actually a very common complementary technique to OKR. So you can imagine, you know, that this team could create an OKR here and let's say the OKR, they wanted to give better project suggestions. Maybe their idea was to build this, you know, goal capturing flow, but importantly, the key result goal capture completion rates matches, sorry, that one. So this KR matches capturing key goals. This key result matches project matching success rate. So what you can see here is that you can create goals for time-based goals. That doesn't mean that you can't use the North Star framework. These two things work together as you're working through. Now, to make this even more real, you know, you could imagine for a second, you know, how would a team work with this?

15:55 So we might have one team. So let's say we have a team called the Hex Pistols. And they're going to come to work for the next 12 months or 18 months trying to improve that project matching success rate. They pick their goal, which is project matching success rate, as we just discussed. And then... We've got their OKR, say set an OKR every particular quarter, and this could be what their work looks like, right? It's a series of ideas here that they're going to focus on, and then they're going to try many different things. So even the language here is important, right? Instead of to do, doing, and done, we have focusing and review, and then to try, trying, and review. And then obviously it would be helpful for them to have dashboards and other things. This is amplitude at the moment, but this is kind of what the team would think about for maybe even years on end, right? They would just keep working this motion. And so you see, again, how this works is that the

17:01 This team, the Hex Pistols, is thinking about these ideas over here and trying different things, but this input does not go anywhere. So hopefully that's interesting or clear. Let me give you another example here that maybe this is from FinTech, but it might help you even understand the concept even more. And what I'm trying to do in this particular example is show you how the team arrived at their North Star. Different teams come up with their North Star and inputs in different ways. So the way that this team came up with their North Star is first they asked the question, what are the parts of the value puzzle? And so you can see things here like you've got to fund your accounts. You want to be their primary bank. You've got to be involved in savings and acquisition and borrowing. So at first, the first part of the process for the team is very... Abstract and they're just thinking about the parts of the puzzle that work. The second step for this particular team was grouping these things into categories.

18:08 So you can see here, for example, the couple groups here were one, funding. So you have to fund your accounts. This is the same with most fintech companies. The second one, you have to integrate into their financial life. The third, they want to acquire customers at or below market rates. Then you need to help people understand and act. That's another part. And then finally, you want to be a trustworthy bank. So the next part of this process was after coming up with the parts of the value puzzle, this abstract activity up here, what they did is they then started to group these things into particular parts. After that, they really clarified the statement. Now, one thing that's extremely important here that you'll note is that These, before thinking about measurement, you notice that these aren't metrics yet. These are ideas or variables or concepts or dynamics.

19:10 So what this team did is they clarified in English in this case, or it could be in French, I guess. Any language, they clarified in words what they were trying to do. So we want to be their primary bank, and we believe that's a function of the effectiveness of funding accounts, our ability to understand and act on the insights that people have, the ability to integrate into their financial life, our trust and reliability, and then acquiring customers. So the reason why I clarify this is that a lot of teams think that they need to jump to measurement, jump to the metrics, and that's not really the case, right? They really need to clarify the statement. Finally, at the end, we won't go through all of these, but finally, Finally, at the end, they clarified what their specific metrics. So we are their primary bank, number of users with direct deposit activity and spending activity during a period of time, funding effectiveness, how quickly could they get money into the account?

20:17 So what was that median time and so forth? So hopefully this gives you an idea. You know, when you look at a different teams, you know, this is one method for coming up with it is to take this abstract group, clarify, and then finally get the measurement correct. Awesome. So what does this, you know, what does this let us do when we're working this way? So what... The North Star framework does is gives a central model for teams to work on. So imagine in the beginning, you know, you've got teams that work maybe as a feature factory. So their job and the company's build this, they deliver, build this and deliver. And the problem with working as in a feature factory is that we don't really think about outcomes.

21:18 You know, output is just the primary way that teams work. Now, what you have next is you have this idea that Teams get these goals, maybe they're working on an OKO. But even this particular method makes it hard for teams to see the big picture. They don't see what they're driving towards over the course of maybe years. They just think, okay, I've got this project. Let me connect a goal to it. And that's how we'll do it. So what we're doing with the North Star framework is that we're trying to create this persistent model. So you create this North Star in the middle. You got a pen there. You create this North Star in the middle. And everything inside this circle. Persists as long as the strategy persists. And then you have individual teams, you know, so for example, you've got a team up here, like the one in our example, they're making experiments, and then they are setting these goals as they make them.

22:19 But that work is meant to impact the inputs of the North Star. So what you get when you work this way is a much more sort of stable approach to working. And, you know, I like to say that, you know, the feature factory is one way. This bottom thing here is what I would call like driving in data circles. You see this a lot with teams that they're very data driven, but they don't have a sense of the strategy. They don't have a sense of the why of what they're doing. The North Star framework allows us. To take both a data-informed approach to what we're doing, but also to think long-term about strategy, which is sort of sort of what this talk is about is making strategy more accessible. So whenever I present this, you know, someone will say, Okay, that's great. Does our company need more than one North Star?

23:22 You know, how could we ever have one North Star for our company? Now, it is true that if you're in a pretty big company, you have different products. And if you had different products, that would probably be a good signal to you that you might need different North Star metrics and inputs. What I try to explain to teams, though, is... Challenge that assumption, right? Because if you, what happens in a lot of companies is that there's basically like shipping the org chart, which is, oh, well, we've got the teams organized this way and each one needs an input or its own North Star and therefore we need many North Stars. Or, you know, in some companies, everything becomes a product. Well, we've got one team and one product, so everyone needs a different goal to be able to shoot for. These problems kind of go on and on. Some companies, they lose sight of their customer. They lose sight of the customer journey. You know, there's a lot of dysfunction can happen here. So what I always remind teams about is that you want the structure of your company to match the strategy.

24:30 and then create impact and that informs your strategy. You don't want the structure of your company to drive the strategy. So when teams come to me and say, oh, should we have more than one North Star? The first thing I do is say to them, well, I'd like you to try thinking about one North Star instead of inputs first and see if it makes sense. And if it makes sense, use that. And if it doesn't, then consider multiple North Stars. But this is a question. that always comes up and my basic answer to this is challenge yourself to see if it's possible to come up with one North Star before you start dividing it among every team. Another question that often comes up is around innovation, right? So someone will look at the North Star framework and say, oh, but we have an innovation team. Do they need another North Star? And again, what I say is, no. Why do you need another North Star? Oh, because they're innovating.

25:32 And so I use this example. This is kind of a silly example. But if you think about it, let's say we are thinking about watering plants. You know. Sometimes we might use a watering can, sometimes we might use a drone. Both of those things are meant to water the plants. Some things are maybe more risky, you know, like the drones. Some things are safer, like the watering cans. But it's not necessary to create different North Star metrics for innovation. The main thing is that some ways you'll have safer, and in other ways you'll have riskier ways to be able to impact the North Star metric. So before we wrap up, you know, and, you know, I'll be there and can take questions. I'd love to answer any questions about this is at the highest level. What is the North Star framework? So the North Star framework is. A model. That's it. It's a model.

26:33 And you could think of the North Star framework and the inputs as a hypothesis. So here we've got our North Star. Here we've got our inputs, right? This is our North Star. This is our inputs. And in the beginning, when we're starting a company or have a new product, we might not be very certain about all the connections here, right? We might, you know, this is just a hypothesis. We don't know yet. This is just a hypothesis. We don't know yet. And over time, what we do is we increase the certainty that we have about this. So this is an important idea to finish on and to think about because one of the biggest anti-patterns, I've done a lot of workshops now for the North Star Framework, one of the biggest anti-patterns that comes up is that a team will say to me something like, We can't decide on a North Star because we're not 100% certain.

27:38 We're not 100% sure that this is the right metric. And I'll ask them, you know, what do you believe, right? What's your strategy? I mean, you have some beliefs over it. And this is where it gets interesting because some teams will say, well, we just don't know. We don't have a strategy because we don't know. And other teams will say, well, you know what? We're reasonably confident. There's some open questions, but we're reasonably confident that this is the particular way to go. So the important thing is to get the beliefs out of your head into a framework that you can use. And then over time, increase your certainty. So, you know, imagine, you know, over time, we're going to become certain that this is the connection between things. Or over time, we're going to understand that this is this thing. And even as it relates to revenue. So some people say, well, we can't prove for 100% that the North Star is driving the business benefits and the revenue, which is up here.

28:45 So again, over time, you might learn, and you might also learn that you're wrong, right? Maybe you learned that that's not what you thought it was. So that's what I'll leave you with is that, you know, in closing, it's just a model. All models are a little wrong. And you don't need to be 100% certain about the model before jumping in. So, actually, at my job at Amplitude, I wrote a book about this called the North Star Playbook, and you can access it here at amplitude.com slash North Star. We go through all the various steps into thinking about it as your strategy. We talk about anti-patterns. We talk about how to fit it in your day-to-day work. And I'm very excited to be, you know, I will answer some questions now live, and that will be a lot of fun. And I hope this was interesting to you. Thank you so much.