Jack Litchfield is Head of Product at Super.com, a travel, fintech, and earnings company with a mission to create more opportunities for everyone to put more money in their pocket. He started his career in film and sound design for horror films before getting the entrepreneurial bug and starting his first company in the grocery space. Since making the switch, he’s worked in product lead roles in broadcast television, esports, hospitality research, parking and micro-mobility, and more. He previously led the product team at Extra, a debit card focused on building credit.
In our conversation, Jack talks about Super.com’s philosophy of experimentation, which results in constant, rapid learning across the organization. He shares his leadership methodology of “leading with a decentralized command,” as well as how making big mistakes often yields more insights than big wins.
One of the important things about rapid experimentation isn’t just to experiment, it’s to rapidly learn. We want experiments to win, like lifting retention by X or conversion by Z, but what I really push our team on is the concept of truth-seeking — where it is better to be proven wrong than to appear right.
Oftentimes, we jump to celebrate something too early or are too critical without looking at all sides, and the idea with truth-seeking is to do our best to get a myriad of data points across different types of data. That way, we build confidence in a claim or a decision. We care about what is objectively true.
There’s very low ego in the Super.com building because you’re often wrong a lot when you work here. People are given the authority to challenge you. I’ve had level one and intern teammates challenge things that I’ve said in the spirit of truth-seeking, and they’ve been right. If you don’t create a culture where you can be wrong at the top, other teammates don’t want to be wrong across the board. I’m wrong a lot.
In tandem with that, rapid learning is our superpower here at Super.com. I’ve never worked somewhere where we’ve experimented and disseminated learning this fast. There’s always a shadow side to moving fast, but we launch at least one experiment per day here at Super.com. It’s incredible.
It’s tough. It’s easy to think that there’s a magical framework that’ll answer all of our questions, but that’s just not the case. However, there are always good disciplines to follow, like trusting a teammate’s gut. It is impossible to be perfect and there isn’t a one-size-fits-all solution, but there are a few things that we can do that can help us tactically.
One of those is that you have to accept that in cases where you are trying to learn quickly, quality will be impacted. We have a concept we call seeding and nurturing. Picture a farmer in the field scattering seeds. They’re not really trying to plant them with tender care and nurture in every single tiny seed because most of them won’t sprout. You will slow down learning if you over-invest in the seeds.
Instead, you need to maximize your volume intelligently. Don’t throw seeds into the gutter, but rather, plant them intelligently and often. Then, with the ones that start to sprout, really nurture those. You have to build a roadmap where roughly one-third to one-fourth of it is undefined nurture time. You won’t know what’s going to go there yet, but it will be something that you planted earlier that you now want to develop.
The second piece is that for products to really win, they have to crush business goals, deeply meet customer needs, but also be magical. You need to take some time to invest in that magic. Team leads have to have at least 10 percent of your sprint working on something that is magical.
One best practice is having a team that you really trust. Part of it is having leaders make decisions, but giving them the space to make mistakes. As long as they’re learning, showcasing those learnings, and moving forward, that is actually sometimes a much better outcome. That’s piece number one.
The second piece has to do with the nature of how fast we move. You want to make sure that the team has metrics they can tangibly move and get signal on fast. And on the flip side, you want to create space for a holdout group. For example, we launched gift cards and released this to a market we didn’t intend to. Originally, we thought of shutting it down, but we decided to leave it on for a holdout group to watch what happens for a year. Though there’s space to move quickly, you do want to watch the group over time.
Finally, and I can’t believe I’m going to say this, but you can be too data-driven. You can talk yourself out of your gut instinct because you are overanalyzing a single data point without utilizing all of the info that is pooling in your head. I like to teach teammates that your gut can actually be very data-driven when you’ve built an existing habit of reviewing the data.. You’re picking up subtle cues that your experience has trained as indicators of success or failure all along the way.
We learned so much more from big mistakes. When that happens, it gives you time to reflect. Disappointment is a powerful emotion that teaches you how to improve. If you think you’re hitting home run after home run, you’re never going to work on developing other skills like your sprint speed running around the bases; you’re never going to become faster because you think everything’s great. But if you hit ground balls or keep getting out and look objectively at how you are performing, you’re going to try many different tactics to get on base and keep moving. That’s going to make you a better player.
A great example of this is the challenge we face with customer retention. Some card types and payment methods fail when it’s time to pay for the next month’s membership, so a lot of energy has gone into different tactics to retry the card. When should we charge it? What should the cost be? Can we lower it? Is this a good or bad thing for the customer?
There are all these questions in this area, so, one of the experiments we ran was based on looking at two models of similar products. So we built both models at the same time to compare them to each other.The idea was to double our learning and take the best out of each experiment to learn as much as we could. It made a lot of sense on paper — we were going to learn quickly and have multiple models to see how things perform.
The problem was that, with too many variances in such a complicated process, there were tiny mistakes in each of them. That happens typically in most experiments, and it’s usually fine, but when those mistakes started to collide in real time and compound on top of one another, some customers started experiencing payment issues, which is an absolute no-no. For instance, sometimes the credits that were supposed to count were not counted or were extremely delayed. Those are just unacceptable losses. It’s OK if we run an experiment and I accidentally break conversion on some piece of the travel funnel, but when it’s the customer’s money or loss, you can’t make that mistake.
A lesson from this is to keep it simple and secure when you are touching someone else’s property, especially your customers. That’s very obvious to say out loud, but it’s a very important theme when you’re trying to move fast. You shouldn’t move fast without extreme measures for double-checking and simplicity.
I believe that a well-powered team can accomplish anything. One of my hobbies outside of work is endurance events and team-based events. I’m so attracted to the power of what a group of people can accomplish that an individual cannot. Time and time again, I’ve watched a superhero-level PM come to our team and think they can do it all by themselves, and they’ve been outperformed by a group of interns. The power of a team is just exponentially better, and I want this to be contagious. I want it to be something people celebrate in the building.
You also have to let them set their own goals that ladder up to the team goal. If I were to come on a call and say, “This is the goal,” that’s not setting you up to be autonomous in the future. So, we have decentralized command as a key thesis to how our mission-aligned teams (MATs) operate. We have a single accountable owner and then enough staffing on that team to execute according to a mission.
A mission for a team might be improving how customers earn, how they navigate through our travel funnel, or how they get cash advances into their pockets. You’ve got to give them the space to make their own calls on what the OKRs need to be. If we do monthly OKR reviews and it’s just me talking, it robs them of owning their results. They need to own their wins and losses and be able to speak to both with equal amounts of intention.
Other than rapid experimentation, our other superpower here at Super.com is that we can move unbelievably fast with a lot of autonomy because I am not a gatekeeper. With 80 percent of the decisions made in the building, I have no idea what happened because I don’t need to know. I just want to build teams. And after that’s done, they are the ones who create the wins.
One of my really good mentors taught me about the player-coach model. The idea is that a player on the field is responsible for scoring goals, while also making tactical decisions in real time. They are the ones braving the “in-your-face” realities and have to make tough calls quickly. The coach, who does not step onto the field, intentionally should be looking at the bigger picture of this game and the whole season. They might say, “I’m just a coach. You’re going to make the call and own the outcome, but here’s what I’m seeing from the sidelines, and here’s how you need to be better.” It’s a really effective way of telling your boss to stop micromanaging by simply asking “OK, who is the player and who is the coach here?”
Secondly, as a leader, you need to take a vacation. It’s a good way to see if the team falls apart when you remove yourself from the building. I think there’s a lot of insecurity when you take a vacation — especially when you come back and everything’s fine. My initial thought was, “Shoot, am I irrelevant? Am I not needed here anymore?” But it’s a sign that you’ve set up an autonomous team. You can step aside and they will keep on working.
The last thing I’ll say about creating autonomous teams is the concept of extreme ownership. This concept was created by a thought leader that I admire a lot, Jocko Willink, who’s a former Navy SEAL commander. Extreme ownership means that everything on your team is your fault. If the PM on team seven, who is four reports away from me, makes a big mistake, it’s my fault. If the CEO makes a wrong decision, it’s also my fault. When you bring that attitude to what you’re doing, it becomes actionable. I can then say, “How could I have helped the CEO? What research did I miss? What challenges did I not tell them about?”
When you approach from that angle as a leader, it becomes contagious. You create a team that has everyone’s back at all times. Yesterday, I was about to bring the wrong research to a call. One of my L2s called me and said, “I happened to be reading your deck because I wanted to help you prepare for this win, and I noticed that you had this mistake in here.” That was them actively saying, “If Jack makes a mistake in this big presentation, it’s my fault. So how can I help?” That’s how it becomes the culture of your company.
It helps tremendously. The people who build these items are vastly more important than how they build them. It is so much better to build a team that will tell me I’m wrong rather than deliver the thing I want. It’s so much more fun when a team is harmonized, when they really care about each other, and when we celebrate each other’s wins and take accountability for each other’s mistakes. There are some comical moments where everybody will take ownership of something.
I should also say that if I am willing to move employees, I should also be willing to move myself as the team changes. Leadership is not always about being at the top and in command, sometimes great leadership is knowing when to follow and when I am no longer the best person to be in charge. At Super.com, we’re bringing in somebody who is going to be my new boss, who is vastly more experienced than I am for where we’re headed, because that’s what the team truly needs. And my turn at the helm is changing as I dive into starting up a new line of business in the company, which is where I’m most needed (and candidly, incredibly excited).
It’s an odd metric for success as a leader, but if you become almost replaceable by the people you are training, that’s a win. When you arrive there, you need to be willing to let a teammate lead and demonstrate to yourself what being a great follower looks like. I like to tell people that leadership is really hard, but so is following. We always want to be praised as a great leader, and leadership is an important characteristic. It’s important to also be a follower sometimes — being willing to adapt and evolve so that the team wins.
LogRocket identifies friction points in the user experience so you can make informed decisions about product and design changes that must happen to hit your goals.
With LogRocket, you can understand the scope of the issues affecting your product and prioritize the changes that need to be made. LogRocket simplifies workflows by allowing Engineering, Product, UX, and Design teams to work from the same data as you, eliminating any confusion about what needs to be done.
Get your teams on the same page — try LogRocket today.
Want to get sent new PM Leadership Spotlights when they come out?
Unlike generic prioritization methods, customer-valued approaches place user and/or client impact at the heart of decision-making.
Cristina Fuser, VP, Product at Buzzfeed, talks about how being tech-led and data-led can be a strategic advantage.
Suzanne Ackerman talks about how she leverages visual mapping to align teams and break down complex projects.
Companies don’t agree on the definition of a product manager. However, the essence remains to drive value for customers and the business.