Today’s memo is sponsored by Stytch:
Want to boost conversion and drive growth? Stytch makes user authentication and onboarding seamless and secure. Stytch offers customizable, out-of-the-box authentication with magic links, one tap social logins, biometrics, one-time passcodes, session management, and more.
With their API and SDKs, you can improve user conversion, retention, and security, while saving valuable engineering time. Current customers have seen a 60% increase in sign-up conversion after spending just one day on the integration. Check out Stytch today.
I’ve always been curious about how people ramp up quickly on an industry and form a perspective in the midst of ambiguity, especially with new technologies such as crypto or self-driving cars. They connect all these seemingly disparate data points, and form a narrative where they see something others do not see. Then, they translate that into some actionable insight and take people on the journey with them – from starting a new company, launching a new product, getting a new job as a PM or expanding their scope as a PM.
Here are some examples:
Google’s bet on the future of advertising
Look at the amount of time people spend on the internet – 50% of their time 10-15 years ago, yet only 1% of the advertising budget is going to the internet, versus 99% is still going to TV. Naturally, the system has to correct itself.The future of currency
Numerous financial institutions failed across the world, and had to be bailed out by governments at the expense of their taxpayers. This underscored the fragility of the modern financial system, where the health of our monetary system is reliant on financial institutions that we are forced to trust to make prudent decisions with the money we give them. Too often for comfort, they fail to carry out this fiduciary responsibility to an adequate degree. (via Ben Yu)The future of experimentation
There’s a huge market opportunity in tools for running experiments. Big tech companies like Facebook, and Google have state of the art experimentation tools. Everyone thinks small companies operate in the same way and they can use tools like Optimizely, but this is far from the truth. Optimizely and other off-the-shelf tools were designed in a world where companies wanted to run experiments that consisted of simple client-side changes (e.g. color A vs color B) but nowadays the experiments they want to run are very complex. (via Davis Treybig)
This type of thinking is really powerful, because if you can form an opinion that no one else has, and develop enough conviction in it, then you can make bets that are almost guaranteed to pay off.
Here are guiding principles to craft such a bet:
Step 1- Remove the mental block. There is no answer.
One of the biggest blockers to having a strong opinion is feeling like there is some singular right answer you are trying to find. This makes it hard to navigate a space because you feel like the answer is out there, and you just have to find it.
There is no right answer – in fact, there are many ways to make money. Consider the top 5 richest people in the world. #1 is Jeff Bezos. You might expect all the other people to also be in tech, but up on the list you also have the Waltons (of Wal-Mart), Bernard Arnault (of LVMH luxury goods) and the founder of Zara – a company that has nearly the opposite play of LVMH’s luxury brand.
Investors don’t base their decisions on reality, but rather on perceptions of reality. Reality is the second best strategy.
Your role is to invent a reality and connect data points.
For example, I have two friends who run investment funds with perspectives on crypto that are both valid, even though opposing – and there are opportunities on both sides:
“Crypto is all spread betting and will be regulated away” -> stay invested in traditional businesses, where there is a lot of opportunity
“Crypto is 90% speculation, but that’s not a bad thing and there will be some great companies that come out of it” -> start a crypto vc fund, take many bets, and find the 10% of companies that will survive
To reverse-engineer one of these world views, you can do 3 things:
Notice your instincts – what are you sensing?
Choose the instinct you can’t stop thinking about, and assume that it is already true.
Find the evidence that points to your instinct being true.
But how do you even arrive at an instinct to begin with if you’re exploring a new space?
Step 2- Explore the dark mansion
This part is frustrating if you think about it from the perspective of “What are the places I should be digging deeper into research? How do I know these are the right places?” I’ve found much more success if I let go of the pressure to do anything useful with my insights and start by taking in everything that’s going on and enjoy exploring the dark mansion.
(1) Draw a map with all the information I collect
A few months ago, I was researching a cryptocurrency ecosystem called Terra. As I was researching, I was pushing new items into my mental map, “These are some payment companies, these are some infrastructure-related companies, these are some things people have mentioned they’re worried about on Discord, and there’s the Inter-Blockchain protocol, and so forth…” I was painting a picture of the landscape and synthesizing a lot of information, but it isn’t very organized — this is expected.
(2) Experience the trough
The trough feels like a random walk. You feel like you are not really getting anywhere because there’s no story. It feels like a dark mansion. You enter and you’re constantly bumping into things in the dark and thinking, what is that?
(3) See a way through
And you slowly start to make out the shape of things. You notice something you just bumped into is soft, and curvy and start feeling around and realize it is a couch. Over time, you make a more coherent map of the dark mansion.
(4) Experience the second trough
Then, you hit the second trough. You think “Ok, I understand this system. But it feels unsatisfying. All I have is a summary of the space. But nothing feels novel or inherently surprising. I’ve drawn some conclusions, but I can’t figure out what to do with this information.”
Then you move into the next phase:
Step 3- Acknowledge the dark mansion is actually an open garden
This is a powerful way to think. It’s easy to get depressed if you see things as a closed system, i.e. Everything that exists is what I see in front of me. It makes it hard to see a path forward — because you see it as a closed system, there are no unknowns to be discovered. On the other hand, an open system has no bounds.
Average product managers see the world as a closed system.
Seeing the world as a closed system makes it hard to arrive at anything novel, and thus see any opportunities. For example, Let’s say you are thinking of building a fintech product to help people invest better, you’d start by understanding the history of investing, the players in the space, and how they got there (from traditional wealth managers to platforms like WealthFront and Robinhood).
At first, there will be nothing novel you discover – but this is the natural process of research — first taking in all the information and synthesizing. The next part is drawing conclusions.
Great product managers see the world as an open system.
Seeing the world as an open system allows you to form a vision about the future. Let’s use an example from Titan, a platform that aims to give everyone access to elite investment management (Apple Podcast). You might have seen their ads.
Here’s an example of how, when you have an open system like the future of investing, you can paint a vision. Ask three questions:
(1) What is the ideal state of the system – in this case, investing? Why is it so exciting?
The ideal state is that anyone can invest – if people are in their 20s, have a risk appetite, and a financial cushion, they should also be able to access the type of investment vehicles that only wealthy people are able to access.
(2) Talk about the current state of the system, and why the dream is not yet realized.
Most investment vehicles that the wealthy have access to are limited to everyone else, who can only invest generally in ETFs and stocks. Everyday people are unable to invest for (1) legal reasons - investor protection saying you need a certain net worth to make certain types of investments and (2) economic reasons - from a fund manager’s perspective, it’s more cost effective to collect one $10m check from one wealthy individual, than thousands of $100 checks from many.
(3) What are all the features that a robust solution needs to succeed? How does it tackle the above problems?
Fund manager’s ability to collect many small checks without the cost of collecting them
A product that navigates legal barriers
A level of services that someone investing a risky asset class would require
Notice what pieces you are interested in, because there is often a reason why you find something interesting. Don’t neglect your intuition.
Treat your opinions as a model you’re constantly updating
Going through this process might be difficult at first, but it can be helpful if you see it as a process, where you’re constantly probabilistically updating your view of the world as you gather more data and talk to more people.
This process may appear slow and intellectual at first, but the alternative is being pushed and pulled by the opinions of others and never taking the time to synthesize your own.
Synthesizing signals you see to find and defend your opinion will pay off in the long run, because you’ll be able to connect dots that others don’t see. And connecting these dots will allow you to create more value as a product manager, whether you're an interviewing for a new job in a new industry or you’re a product manager looking to expand the scope of your work and need to convince others on your team to enter a new area.
This is a very helpful post for crafting vision and doing Needle moving work. However I felt that the example/steps you've taken to decompose a new area and predict its future could've been more explicit and clear. Can you share a more detailed example, which covers the actual notes you've taken during all these stages to finally converge into a vision doc (please also share this doc)? That, would make things very clear