Nov. 4, 2023

Summary: Lessons from scaling Stripe | Claire Hughes Johnson (ex-COO of Stripe)

Summary: Lessons from scaling Stripe | Claire Hughes Johnson (ex-COO of Stripe)
“I would say, advice to people generally, is be a force for positive momentum, and it will be actually a real career maker.”

Context

  • Claire Hughes Johnson was most recently COO at Stripe for the past seven years. Before that, she spent 10 years at Google as VP of self-driving cars, VP of global online sales, and director of sales and ops for Gmail, YouTube, Google Apps, and AdWords. She’s been in politics, and is also on the board of HubSpot and The Atlantic.
  • Claire is the author of the book Scaling People, which covers topics like building operational cadence, operating principles, and management versus leadership.
You can also see the episode transcript and Claire’s references below.

Early challenges at Stripe and how Claire started writing Scaling People ▶️

  • Stripe co-founders John and Patrick Collison inspired Claire to write a book after getting a ton of questions about how they scaled Stripe. They needed “Claire in a box.”
  • She also found early positive feedback from her chapter in Elad Gil’s High Growth Handbook.
  • Her biggest lesson from working with the Collison brothers is how much it pays to seek out knowledge from others. “I don’t think I made enough phone calls in my earlier career and asked people’s advice and asked for help and found the person who did that thing five years ago and found out what they learned.”
  • Implementing levels and ladders was challenging but necessary for growing companies like Stripe. “This stuff probably makes people’s skin crawl. You’re immediately getting into something that feels suboptimal, but it is worse to have nothing because it starts to feel very unfair in the environment. And you need it sooner than you think.”
  • Rolling out 24/7 support in multiple channels and languages was very challenging for a high-scale B2B company like Stripe, but essential since they’re in payments.
  • The best advice she received was going faster on some of her intuitions about using outside vendors and outsourcing parts of the model. It was hard find the right balance, but it needed to be done.

The lack of job titles at Stripe ▶️

  • Job titles signal hierarchy and authority, which gets in the way of a collaborative culture. “If you’re the person who has the knowledge, you are the expert on the thing, you better be in the room helping make the decision and then helping drive the decision. And I don’t care how senior you are.”
  • Titles are especially problematic in an early, growing company, where you have to spend increasing amounts of time playing title Tetris to keep everyone happy. “Very soon you have seven VPs in a 25-person company—does that really make sense?”
    • But when engaging with external partners or customers, it’s essential to present your scope of responsibilities when titles are lacking.

Should founders focus on PMF or organizational structure early on? ▶️

  • Product-market fit is the most important, but don’t get too far ahead because the product isn’t the whole company, and neglecting it can actually harm the product itself.
  • Coming into Stripe as an outsider made it easier to identify that the company needed to scale. She looked at the numbers, the inbound support demand, the inbound sales leads, and it was clear the company needed to be twice its size. And not just sales—many other functions like risk, compliance, data, etc.
  • You don’t want to mandate a lot, but enough that everyone can play with the “stack” of the company. If you don’t start putting those things in early, people will just invent those things—picture a house that got worked on 17 times and it’s not even two years old.
  • Hiring
    • A lot of the story you tell your investors, you should be telling internally to anybody you hire. You can get pretty far with that content.
    • Don’t just hire people who are your friends; think about the actual capabilities you need.
    • Evaluating someone accurately is critical, but it’s hard to do that in 30 to 45 minutes, so get good at that.

Personal operating principles and values ▶️

    • It would be easy to build companies if there weren’t humans involved, but there are humans and they’re complicated. A lot of people think management starts with the team, or the product, or the company. It actually starts with you. Self-awareness is the first operating principle.
    • Values exercise: Her book has an exercise where there’s a menu of 70 to 80 values and you force yourself to pick three that matter to you. These usually come from the most important or impactful stories in your life.
    • Assess your work style tendencies using tools like Myers-Briggs or DISC. A more extroverted task-oriented person might be a good director, or a more extroverted people-oriented person might be charismatic and great at selling a vision.
W230227_JOHNSON_WORK_STYLE_360.png
  • A personal operating principle for her is saying the thing she thinks she cannot say. Often your biggest strength is something that you don’t know is a big strength because it’s almost like breathing. For her, saying something fairly openly and directly but in a non-threatening way is that.

Advice for saying uncomfortable things ▶️

  • Ask a question. A question is not threatening. It could be as simple as “Is there something we’re not talking about? It feels like to me...”
  • And then own it. This is your observation, your perception. This is not a judgment. “Lenny, you know what? I wonder if you missed an opportunity in that interview. Did you feel like you missed an area that...”
Just doing these two things will get you way further than you would’ve ever thought in sharing. There are likely other people who also want to say the same thing, and you’ve just opened up a door that they can walk through.

Be an explorer, not a lecturer ▶️

  • As a manager, your job is to enable people to be their best, not be the expert and tell people what to do. Most of management involves exploring with someone rather than teaching them. “Intuition” as a word gets a bad rap, especially if you work with a lot of engineers, but what is a scientific hypothesis if not a well-informed piece of intuition? Too many managers wait until they have a million pieces of data. Go earlier—all you’re doing is holding up a mirror, and you have to own it, e.g. “My experience of you in that meeting was that you seemed nervous to me. Maybe you were not, but actually maybe this is just a physical coaching opportunity.”
  • Reid Hoffman, founder of LinkedIn, who admits to being a leader not a manager, emphasized the importance of creating an environment of open feedback. One of his employees once told him, “Reid, I wouldn’t trust you to manage a McDonald’s” and was perfectly comfortable doing that.

Touchstone documents and operating system ▶️

  • An important role of leaders is to create stability in a chaotic and high-growth environment through rituals and common practices, e.g. quarterly or monthly goals. Talking to people in similar situations will help you realize that most places are crazy and chaotic internally for a long time, and that’s more normal than you realize.
  • Codify these operating principles into a set of “touchstone” documents that you can come back to to regain a sense of control when things get overwhelming. See Stripe’s example here.
  • Like a house, your company needs three things:
    1. Foundational structures (next section)
    2. Supporting structures: How you do things, e.g. quarterly business reviews, OKRs, planning processes
    3. Mechanicals: Your operating cadence or rhythm, e.g. an internal demo event six months before the annual conference to pick out interesting ideas that can be launched

Founding documents you need ▶️

  • A clear mission statement that represents what your startup is trying to accomplish. For Stripe it was “increasing the GDP of the internet.”
  • Long-term goals that provide more detail on your mission, e.g. one of Stripe’s is to advance the state of the art and developer tools. Customers may not care as much about that, but their forever user is fundamentally a developer and they care deeply about that experience. Long-term goals should guide almost all the choices and actions at the company.
  • If you’re having trouble writing your company goals, zoom way out and ask, “If we’re going to meet this mission, what do we have to accomplish in the long term?”

The components of a company’s operating system ▶️

  • Key questions:
    1. What is your structure for setting goals or milestones?
    2. How do you measure progress on those goals?
    3. What’s the cadence of your processes?
  • On cadence:
    • Play around with the time frame, and don’t feel restricted by what other companies do. Stripe changed their “quarterly” business reviews to every six weeks.
    • Signs that your cadence is off: not enough progress being made between reviews, stale content, or the feeling that the meeting has already happened.
    • Some companies will have a pretty frequent metrics review cadence and then an infrequent strategy review cadence. What happens is the metrics review becomes the strategy review, because you’re not talking about the strategy often enough.
    • Don’t fool yourself that having more frequent checks on things is going to actually make things run faster. The opposite can be true.
  • At Stripe, if they’re looking at metrics, they look at their dashboard live, not a special presentation. That way you get a really good dashboard that’s real-time, and nobody’s wasting time prepping all that information. They took it a step further and randomly chose someone to present your data at the metrics review meeting (they called it “spin the wheel”).

COOs and which types of businesses need them ▶️

  • Most companies do not have a COO (more than 80%), and it’s not an automatic role to hire. It can be useful in high-growth companies, particularly those with a deeply operational or manufacturing component, like Apple, or Google’s hardware businesses.
  • Between achieving product-market fit and having to build a company, founders can have a lot on their plate. That’s where a COO role can give you leverage. But note that there’s no mythical silver-bullet COO hire or panacea.
  • Hiring a COO: A low-risk strategy could be to bring in, say, a head of business operations or a role that’s sort of COO-like, and try a couple of functions. If they scale with the company and with the role, maybe you can decide they’re a COO.
  • A fantastic COO doesn’t mean everything’s hunky-dory. It means just the right amount of friction and tension with forward momentum and mutual trust with the CEO, enabling the whole to be greater than the sum of its parts.
  • Pitfall: Some founders think a COO would just solve all the cultural issues and let them somewhat shirk the responsibility of a founder. They should really be asking themselves if they like running a company at all.

Internal communications as a tool for scaling ▶️

  • Have common codified documents of what you believe in, how you work, and then repeat them till you’re blue in the face. Internal comms are not extraneous.
  • Use tools smartly, e.g. an intranet site, newsletters, and messages or videos from founders.
  • Maintain a stable cadence.
  • A good internal communication strategy should be replicable across teams.
  • Talk to sales leaders to gain insights into keeping everyone on the same page, because that’s a big part of their job.
  • Use offsites to create cohesion and alignment in teams. “You're basically activating new parts of their brain, and then you’re also having a group experience that cements a belief system usually, or a set of plans.”

Running effective meetings and making decisions ▶️

  • Claire’s video on running effective staff meetings
  • Number-one thing: get rid of the meeting if possible. The barrier to creating meetings is too low. People need to know why they’re meeting (make a decision, share information, etc.). You need to know who needs to be in the meeting, who’s making the decision, and who doesn’t need to be there (because you can just send them a note afterward about what happened).
  • Make decision-making explicit using frameworks like SPADE or DACI, clearly defining who’s making the decision and the criteria used. Follow the model and do it—you can always revisit the decision, but don’t get paralyzed.
  • If you’re not sure who the decision maker is, it’s probably you. And if you don’t act that way, you slow the whole company down.
  • Bezos framework: irreversible vs. reversible decisions. Make the latter quickly.

 

This is a human edited summary of the podcast episode with Claire, by Gaurav Chandrashekar (@cggaurav, productscale.xyz). To listen to the full episode, go here.