Leveraging Automation & Rituals To Fuel Growth

at UXDX Europe. Video: https://youtu.be/rpIULgmxrf8

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Waking up and defining terms

[00:00:01] Morning. I was expecting a smaller crowd, being the first person of the day, but it's really great to see you. First, thank you for not partying too hard and coming in today. I've got more than 30 slides and 30 minutes, so I'm going to be going fast. But before I get started, I need your help. All of you work in product, so I'm going to try something to wake us all up. How many of you do a daily stand-up in the office? All right, good number. How many of you end your daily stand-up with a clap? Not many. All right, this is going to be interesting.

[00:00:40] Let's try it. It's simple: I go three, two, one, and you clap. Now, before you do that, I want you to imagine something. Look around you and imagine that you did not party last night, and everyone else here has partied really hard and needs to be woken up. They depend on you. Shall we try it? Three, two, one, go. Just one. I should have mentioned this: one clap. But that's good, we're awake.

[00:01:13] Let's start. What is my talk about? It sounds a little dry when you say rituals and OKRs and automation, but it's pretty simple. In today's talk I'm going to be talking about how you can use automation and rituals to power growth, and by growth I mean the outcomes that you and your company want to achieve.

[00:01:36] What is automation? First let me just define that. As soon as I say the word automation, all of us have this association of automation with machines, and that's because for the past century automation has been mechanization, making things into robotics. But that's not what it is in our world of product building and consumer psychology products. Automation here, in consumer psychology, is about making better decisions. Imagine you were a person who had to make decisions, and you had a process through which you could make smarter ones, because your cranial capacity can now be used for other things. That's what automation really is: it's allowing you to make the decisions that help you, smarter and faster.

[00:02:22] The second bit is rituals. This is literally what you need to do to get your outcomes: the execution pathways that you think about day in, day out. I'm going to be covering two parts to this, but I'm going in the opposite order. I always tell my teams this: you can't get a person to go to any destination until you tell them the outcome, because if you don't tell them the outcome and they don't want to go there, they're not going to come with you. So I'm going to start this presentation the opposite way. I'm going to talk to you about Stash and what Stash wants to do as an outcome, and then I'm going to dial it back to the pathway of execution that we have chosen and the rituals that we follow along the way. That's simple.

Behaviorism and the jello fellow

[00:03:10] Before I start, I want to tell you: anyone who builds product is working in a field of behavior that's almost a hundred years old. I don't say this trivially. It's not that you are not innovating today, but literally the field that supports product is behaviorism, and that is almost a hundred years old. When I was in college, I used to study Skinner and his pigeons, and Pavlov and his dog. Behaviorism is a very old field, but a lot of us are simply digitizing experiences that have been well studied. What that means is all of you have access to a wealth of knowledge that already exists that you can take into your products today. I believe behaviorism is where it comes from.

[00:04:03] How does this work in my own life? I thought I'd take an example. This is me, that's my family, two girls. I had a crisis in December. Before I get to the crisis, let me tell you a little bit about myself. I'm a person who hates gyms. I don't like going to gyms. I've gone on and off to many of them, and when I go to the gym, the things I hate about it are lots of them. Remember this gleaming array of big machines, white and black? I literally paralyze myself. I do not know what to do. There's a machine for almost every muscle in my body.

[00:04:39] What I end up doing is going to the one machine that I really know well, which is the treadmill, and I start running. I don't really even run there. What I do is I run, I look across, I see the person running on the machine next to me, I check their little speed, and then I go a little faster than them and try to keep it up. This is why I have never been able to stay with gyms.

[00:05:00] Now, the crisis in December. My daughters looked at me, they looked at my tummy, and they said, "Dad, you are a jello fellow." This struck me so hard: my favorite dessert, jello, being combined with my personality. I said, this is a wake-up call. What am I going to do? So, like all good product people, I got myself an app. It's called Seven [?], and what it did was literally allow me not to have to think about exercise. I dislike exercise. It started me sneakily, saying I could do only seven minutes, and it has gotten me into a routine. I have a routine every morning: at 6:30 I wake up and do it. The outcome was simple: I lost 15 pounds. Now, I know what some of you are thinking, I could lose another 15, but that's another story.

Stash and the American middle class

[00:05:53] Let's start this with Stash. Stash was started from a need in the industry. There are about 100 million Americans who do not have full access to financial services. I don't mean that they don't have a bank account. I mean they're not treated as first-class citizens. It's not that they do not have good financial habits. They have great credit scores, they have a lot of wealth combined in aggregate, and they make a lot of spending happen. These hundred million people spend billions of dollars on financial products, everything from investing to cards to savings. There's a lot of money that they spend collectively, but the industry does not treat them well.

[00:06:44] These consumers hate the legacy banks. I remember the last time I walked into a bank. Literally, banks are going the way of Amazon, if you think about it. I walked in, and there were more people behind the desk than there were in the line, and I said, why would they run this facility? These people do not like their banks, and there's a reason for this: singly, as a single person, they do not rise to the level where they grab the attention of banks. Banks tend to look for people who are wealthier. At Stash, the average demographic is about $50,000 family income. That's not the place at which you get a $350 deposit just for signing up at the bank. They pay a ton of fees, 35 billion, and they literally are not knowledgeable about financial services.

[00:07:39] All of this together comes down to what Stash wants to do. Stash wants to be the financial home for the American middle class. We have about 4 million people on our platform today, and we know our target is 100 million. Did you get that working? We want to have three types of solutions: banking solutions, investing solutions and insurance. If you think about people in their lifecycle, we want to get everyone to investing, but banking is a pathway to that.

From vision to a concrete strategy

[00:08:10] That's all great. So you've got a financial home for the middle class. This is the point at which you ask me the question: what does that mean? Yesterday Cyndi [?] from Mastercard spoke about a vision, and visions are great. I love them. They make you aspire, they make you dream. Stash's vision of being a financial home for the middle class truly resonates with me. I want to make that happen. But that doesn't help me become a good product leader. As a product leader, I need to get to the next step.

[00:08:42] Think about great aspirational brands. A great example is Nike. They spend billions of dollars on their "Just Do It" push. What does that mean? If I were Elon Musk, "Just Do It" might mean: I'm an astronaut, and how do I get a red Tesla to auto-drive me on the surface of Mars? If you are me, "Just Do It" might mean: do I pick oatmeal or a bagel [?] this morning? That's the range of aspiration.

[00:09:11] The first solution to taking an aspirational vision to execution is having a very concrete strategy. In the financial industry, the word product means something very different from what most of you are used to. It means financial products, which means a debit account, a credit card, a mortgage. These are distinct financial products, and that ecosystem gets tightened because you have a regulatory environment around each particular product. So when you talk product in finance, you typically are talking about a specific financial product. This creates chaos for product builders, because Stash itself, as it evolved, became an app of apps, because we were talking about an app of financial products.

[00:10:00] How do you get it back to becoming truly a product? That's to think about the user, the consumer. How do you define a consumer's need singularly, so that it spans all the financial products and all the features that you have? Focusing on the strategy, the first thing is to think of the consumer need. That need should spread across the ecosystem, everything that you do. Some of you may have smaller problem needs, and that allows you to be well defined. Some of you who have to spread will have to be able to distill this down. Second, you think about building world-class solutions against those needs. And third, the business value pops out.

[00:10:43] In Stash's case, it's very simple. All of you have the same need. We all work from paycheck to paycheck. We manage our spend from that paycheck. Anything that's left over, we tend to put away for the future, and then we want to protect our money from risk. That's it. That's the need that spans your entire ecosystem of managing money. So "financial home", in terms of a customer need, simply means that. Second, if you think about this need and how Americans actually stack up against it, you see that they don't do very well. A lot of them, forget multiple needs, are not doing well on even one of them. That clearly creates opportunity.

Tools, guidance and automation

[00:11:26] When you have opportunity, the next step: I mentioned needs, then solutions. In Stash's case, we think about solutions in terms of three things: tools, guidance and automation. It need not be the exact feature you build, but you need a framework. We build tools to help you get started, we give you guidance on why you should be using these tools and what tools you should be using, and we give you automations. I'm going to refer to automation a little later; that stands out for us from the features that you use. It's a core part of the platform.

[00:11:57] Having each one of these, basic tools, guidance and automation, and marrying these two gets you a framework. You think about generation of money, interaction with money, and then protecting your money, and on the right you have the dimensions of tools, guidance and automation. Everything we do is mapped into this. This framework allows us, alongside research, to then go into prioritization: what should we be doing next? If you're building an ecosystem, you need to think of what you're building tomorrow, six months from now and three years from now. More importantly, this framework allows us to not do the things we should not be doing right now. In fact, we may never do them at all.

[00:12:40] That's the second part: being able to map your products. The third part is very simple. You get these two right, you get the needs right, you get good solutions, and you get business outcomes. This is average revenue per user. That's an outcome we all go for; a lot of you are in this industry to commercialize it. So I talked about having a product strategy.

Squads and squad design

[00:13:02] Let's talk about execution. The most granular unit at Stash is squads. I'm sure you've heard the name. We're not fully agile, we don't do everything agile, but these are fully staffed, full-stack squads, with designers, researchers, back end and front end, product marketers and product managers. The thing about squads is you need to think a little more about how they start running. If you think about a team, a guerrilla team, they need to run. How do you get them to run? Where do they run to?

[00:13:33] The first thing to think about is squad design: what is the core of your squad? The second bit I'll be covering is the guide rails, the guardrails so to speak, for your squads, so that when they run fast they don't go off center and they stay aligned. The third bit is what rituals these squads actually need on a day-to-day basis for execution.

[00:13:55] I'm sure some of you have seen this picture; it's a famous one. All the young men [?] in squads are like that. One version of squads is: I think about Stashers, I meet them for the first time, I impress on them what value Stash brings to them, I get them working on the platform, and then I retain them over time. This version of life is the lifecycle view. The second version is that a particular Stasher may be purchasing a stock on a particular day, or perhaps swiping their bank card on a particular day. In that singular experience they need to have end-to-end satisfaction, so the concept of service design becomes important. When you think of squads, you cannot have one or the other. You must have both. You need to be able to think of people in aggregate, because they are human beings, and you need to be able to think of experiences in isolation. So we have two kinds of squads.

Design system and measuring tests

[00:14:52] The second thing: product building is about storytelling. The best definition I've ever heard of product building, and what I personally feel, is that you're a storyteller, and a storyteller is only as good as the tactics and techniques you use to tell the story. Your design system is an integral part of this. Imagine you were listening to a speaker who had a terrible voice and terrible body language. You would not pay attention. The same is true of product building. Design systems give you consistency, they give you predictability, they decrease decision-making.

[00:15:29] We revamped these screens. Before, we didn't actually have a design system; we had evolved and just put together screens one at a time. That happens in startups too. We mapped out a new design system and went with the values that are important to Stash, and they are here: empowered, motivated, calm, guided, safe, relevant. These were the core principles of the new design system. We then tested this design system with customers, they ranked it, and you see the scores. If you ever see the Stash brand, you'll notice the purple color is very strong.

[00:16:10] This is important. As product people, we always build stuff, and a lot of teams will be very happy with an A/B test. Connor from Duolingo yesterday talked about how A/B tests need to be guided: you need a good hypothesis. What I want to call out here is that a lot of the A/B testing tools you use off the shelf tend to be very short-sighted, because they typically look for optimization of a very short-term step: maybe a funnel, maybe the next step of the funnel, maybe a screen downstream. The truth about A/B tests is there are huge opportunity costs for a test that you do. You need to have an aggregate view of what actually happened in the ecosystem.

[00:16:56] So we built out a complete tool called Wisdom [?] to tell us exactly what happens from your test, and that required us to build data science lifecycle value and lifetime value monitors. For every test, you can give the key metrics you are trying to move, but you can also watch almost 30 to 40 other metrics that are changing in the ecosystem. This allows us to not make short-term decisions, to not just move one metric and have an opportunity cost in the rest of the ecosystem. Squads use this too.

Rituals for transparency

[00:17:28] The third bit squads need in any organization is transparency. All of you are in a product team, which means literally the rest of the organization constantly comes up to you and asks, are you doing the right thing? That's why this is so important. We follow quite a few rituals. Some of them are very granular, like daily stand-ups. Some of them go heavier. We have on-demand design studios. They do research, they do low-fidelity prototyping, they do high-fidelity prototyping. That was an evolution. We initially did only prototyping in the field, but we started doing low fidelity, then high fidelity, and then going to the field. It just saves you tech costs. Tech is a very expensive resource.

[00:18:14] We have quarterly OKRs. For those of you who are not familiar with the OKR process, it's just an aspirational objective and a key result that you want to get to. You typically get three of them for a squad, and then you monitor them. That's more important: the difference between SMART goals and OKRs is that OKRs are supposed to be a continuous process of watching. So we have quarterly OKR settings, and then we have monthly OKR reviews. Squads do all this; the teams actually organically do this. Yesterday there was a talk about autonomous teams. It is very important for teams to buy into the direction they're headed, and for them to transparently push out information so that the rest of the company is comfortable with that direction. And we have weekly product reviews. These are demos, and we love these. They create this constant check that the quality of what goes out to the field is acceptable at the Stash level.

Automatic: control, choice and outcomes

[00:19:05] Now I want to talk about one specific thing called automatic. I refer to it as not being robo. I hate the term robo-advisor. It's just a terrible name, because, think about it, no one wants to be a robot. Everyone wants to be in control. So I'm going to talk about the principles of what automatic means. We have a specific squad at Stash called Auto-Stash to create that elevation of automatic beyond just features. The first thing about automatic is being able to give control. It almost sounds opposite: how is automatic giving control? But I'll talk to that. The second is giving choice in whatever they're doing. And I'll talk about Stock-Back, which is our latest automatic feature.

[00:19:58] Talking about control: in automatic, avoid dark patterns. People want to know that they've actually made a decision to go automatic. This was a flow in which people turned on something automatic and did not realize it. It is a dark pattern when this happens. When people go automatic that way, it might feel like you moved a KPI, but it creates a very unpleasant user experience, because when they realize it, they're like, "You tricked me." These are literally the words they sometimes use. So avoid dark patterns in automatic. When people are choosing to make a decision that helps them for the longer term, they need to know.

[00:20:35] Second, you give them control over what they want. There's actually a Stasher in the audience I was talking to yesterday over dinner, and she was telling me about how she Auto-Stashes, and she turned off Auto-Stashing for a month because she had some changes in her life, and she never went back to it. Our research indicated the exact same problem. It's not that people want to stop it; they want to pause it. They just have something going on and they want to come back to it. So you need to give people control over how automatic actually works for them.

[00:21:14] Third is choice. Some of our Auto-Stashers put money into cash. They just put away money as cash; they need liquid money. And some of them put it into investments. Some of them have flexible work; some of them have regular paychecks. Which really means we have different products for these people. Some people can actually put away money every Friday. For some people, we have algorithms which determine whether you have spare cash. And some people just like to save in proportion to spend. These belong to a class of rules called IFTTT, if this then that. There are a lot of them, but we have a smaller set, which is based on user research.

[00:22:02] The most important thing when you're getting people to do automatic is that you definitely need to give them insight into how it works, how the machine works. But the more important bit is they need to have buy-in into the outcome that's going to happen at the end of automatic. An example here is what happens if you put aside $20 a week. Sure, we'll send you emails, we'll send you notifications, we'll help you track it, but the outcome of automatic is literally what drives people into action. As I mentioned, automatic is smarter decision-making, smarter decision-making to achieve some objective. So you start with that. If you're trying to get people to do automatic, you start with objectives.

Stock-Back: you were always an investor

[00:22:41] At Stash, we're really proud of this. We had a whole bunch of people starting with us on investing, but then there were a lot of people not ready to invest. We asked them why, and we were surprised: a lot of people did not even have the concept of a stock. It's sometimes joked about in the US: Main Street doesn't care about Wall Street. The smaller towns in the US all have a Main Street, and they don't care about Wall Street. The amount of financial guidance that's needed to get this demographic to move up is challenging.

[00:23:21] So we decided to think about this differently. We said, all right, you do not know about investing? Are you sure? Let us show you that you were always an investor. The way we did that: we said, all right, we have a debit card, and if you were to use this card at any venue, let's say you were to buy something on Amazon, you would get Amazon stock from Stash. If you were to use it to stream a video on Netflix, you'd get Netflix stock. If you were to use it to buy a doughnut at Dunkin', which is a popular chain in the US, you would get Dunkin' stock, and you would get Dunkin' Donuts, which are like... so.

[00:24:06] People naturally have associations with brands, and we used that association to move them slightly and say, these brands that you love are actually companies, and you can have stock in them. It doesn't matter if you have a hundred shares, one share or a fractional share. The knowledge you gain and the gateway that opens for you is literally the same. So Stock-Back is our product. It's unique. It's a debit card: you swipe it, and you get stock wherever you swipe. It's great.

Results and the Stash community

[00:24:44] I have five minutes, and I want to talk about what this all did. As the expression goes, the proof is always in the pudding. Stash revamped its app this summer. We launched a completely user-centric app in our new design system. You see the purple, and if you look carefully, you'll see an S in the back. I've been told that's not an S, it's a journey, but that's OK. We got 60% of our Stashers to put away money automatically. Why is that not 90? Perhaps we didn't give enough control, perhaps we didn't give enough choice; we haven't understood the others. But 60% of our Stashers have made smarter decisions. They are not as affected by stock market volatility, because you have this whole concept of dollar-cost averaging, and this helps you get there.

[00:25:40] We've given over three and a half million stock rewards through our debit card. That's three and a half million opportunities for people who do not know about stocks to learn about stocks, to stop for a minute and reflect: stocks, what are you talking about? What can I do once I learn the basic concepts of it?

[00:26:03] The thing I'm most proud of is that we have a self-organized community. If you follow the financial industry, you're regulated; you're not allowed to create communities. We have a few of them. The largest of them has more than 70,000 people. It's on Facebook, it is self-organized, and the beauty of this community is that they support each other. These people talk to each other in a language that is so organic that Stash could not recreate it. If you know Humans of New York or similar sites, you'll see people talk about everything. They post a financial profile, they talk about their personal lives, they feel sad when they have to stop Stashing for a little bit, and they get all the support they need from the community. "What do I need to buy?" Ten replies. "What do I need to do next?" Another ten replies.

[00:27:04] The important bit for us is they're funny as they do it. I love the funny bit. They talk about market moments. As you see, I put up one of them. It says, "I love long romantic walks to the bank." I thought it was funny. If I said this to my wife, I'm pretty sure she would not think it's funny, but that's Stash.

[00:27:22] In a nutshell, to recap: when you think about an outcome, whatever outcome you or your company desires, distill it down to a clear strategy. Once you have that strategy, in terms of needs and a framework for solutions, move to the next phase: how do you design execution to be efficient? Once you have both of those, see if you have to elevate anything in behaviors beyond being a basic feature and make it a core part of your platform. That's all I've got. Thank you.