Sunny

← All weeks

What I’ve been reading (W28, 2023)

Lessons from Figma · Product lessons from threads

Lessons from Figma: Being customer-obsessed, service-oriented and building feel-good software Customer support is everyones’ job.

When we launched our first private beta, we were fielding so many support requests that it became a core part of all of our jobs. We would go back and forth with our beta users so much that it was like being in a giant private chat group, talking about this experiment called Figma that we were all building together. And because we were all in it together, it felt natural to drop everything if a bug or usability issue got in the way of a customer doing their work. That attitude soon became part of our early culture: Customer support is everyone’s job.

This attitude wasn’t just about responding to support tickets. We would do things like drive an hour to go visit a customer who was having performance issues on their computer that didn’t show up on ours.

We would respond and chat with someone over the weekend if they were working on an important project and something went wrong. Their problems were our problems.

Your users don’t judge your velocity based on how many lines of code you write. They judge you based on how quickly you do the things that matter.

Figma’s rhythm for customer support (cc: )

Our rhythm was simple: (1) understand the customer’s problem, (2) unblock the customer by giving them immediate support, and (3) if needed, change the product so that other customers don’t run into the same problem. We used this approach for bug fixing, of course, but also to find gaps in our product, which we would address through bigger, bolder features.

Service-oriented mindset

One trick that I use is to think of Figma as if we were in the service industry. I imagine that Figma is a hotel or a restaurant, and our customers are our guests. It’s taking the term “Software as a Service” (SaaS)* and breaking it down, word by word, as if we are literally providing a service to people. We just happen to provide that service by writing software. This is a service-oriented mindset.

The importance of feel

When you talk to designers about their preferred tools, they start talking about how things “feel.” It’s almost like talking to a musician about their favorite instrument. It’s my belief that the feel of software is vital to adoption, especially when you are going for product-led growth. People respond instinctively to how something feels, and you can’t fake it. In fact, I think it might be the most important part of PLG. Cutting multiplayer out of the initial release was really tough, but our thinking was that if we had started with something that felt bad—even with all the right features—we would’ve started with a credibility problem. We would’ve had to then spend the majority of our time convincing people that we weren’t bad anymore. Even though we didn’t have every feature from the start, this trust allowed our customers to believe that we would continue to improve and add features based on their feedback.

The elements of feel-good software

Every product will have different aspects that are important to get right for the feel of it.

For Figma, we focused on the following: Performance—We measure how fast we are able to display changes as the user edits the document. We try to hold ourselves to a bar of 60 frames per second whenever we can. Familiarity—Design tools have a long history of keyboard shortcuts and muscle memory. For example, certain modifier keys affect how objects get resized, and there is certain behavior that users expect if the keys are pressed or released halfway through. There are even combinations you can do, like holding down the spacebar while moving an object, that feel like advanced video game combo moves. Consistency—Figma can be quite complex. Objects can be duplicated from file to file, nested within each other, and updated in real time while multiple people edit at once. We have an internal set of rules for how everything operates, and these rules help people feel like the tool is predictable, even when they can’t quite put their finger on what those rules are.

Product lessons from threads

  • Thoughtfully solving the cold start problem: Threads got everyone on Instagram’s social graph, feed algorithm, followers and even pre-seeded content from influencers before it went live for everyone.
  • Timing matters: Meta actually pulled forward the launch of Threads after some more mishaps from Twitter, where they decided to rate limit users on Twitter and also stop allowing non-logged-in users access to tweets on Twitter.
  • Distribution is king: Threads hit 100M users today, doing so in approximately 5 days. This surpasses ChatGPT, which accomplished the same in 2 months.

Calendar as a time machine (cc: @sung will love this)

  • Time is a curious thing. It’s a constantly flowing stream that can’t be paused, stopped, or repeated. We experience it, but we can’t control it. We can’t even touch or feel it.
  • Calendars, on the other hand, cover the entire spectrum of time. Past, present and future. They are the closest thing we have to a time machine.
  • Calendars allow us to travel forward in time and see the future. More importantly, they allow us to change the future.
  • Changing the future means dedicating time to things that matter. It means allocating our most precious resources to activities with the highest expected return on investment.
  • Treating todos as calendar events is helpful because calendars introduce constraints. A calendar forces you to estimate how long each task will take and then find empty space for it on a 24 hours × 7 days grid, which is already cluttered with other things. It’s like playing Tetris with blocks of time.
  • We only use our calendars to look forward in time, never to reflect on things that happened in the past. We can use calendars to reflect.

Thoughts on code review (cc: ) What effects are code reviews supposed to have on team software development?

  • Reduce behavioural errors.
  • Reduce structural errors.
  • Improve team understanding.
  • Accelerate learning.
  • Reduce team risk.

In what order & at what granularity should developers separate, work, & integrate make changes?

  • Flow. All else being equal, smaller batches of value more frequently are worth more than larger batches of value less frequently.
  • Rowing. We’re all in the same boat, so it doesn’t matter how fast one person can row, it matters how fast we can row together.
  • Overhead. Every increment costs, so have as few increments as possible. (Yes, this contradicts the Principle of Flow. It’s complicated or we wouldn’t see fights about it.)
  • Alignment. Align authority (the ability to take action) with responsibility (consequences flowing towards power).
  • Integration. It’s not divide & conquer, it’s divide, conquer, & integrate. Account for the cost of integration. Including mistakes.

Others

Substack edition

View the original post

Lessons from Figma: Being customer-obsessed, service-oriented and building feel-good software

Customer support is everyones’ job.

When we launched our first private beta, we were fielding so many support requests that it became a core part of all of our jobs. We would go back and forth with our beta users so much that it was like being in a giant private chat group, talking about this experiment called Figma that we were all building together. And because we were all in it together, it felt natural to drop everything if a bug or usability issue got in the way of a customer doing their work. That attitude soon became part of our early culture: Customer support is everyone’s job.

This attitude wasn’t just about responding to support tickets. We would do things like drive an hour to go visit a customer who was having performance issues on their computer that didn’t show up on ours.

We would respond and chat with someone over the weekend if they were working on an important project and something went wrong. Their problems were our problems.

Your users don’t judge your velocity based on how many lines of code you write. They judge you based on how quickly you do the things that matter.

Figma’s rhythm for customer support 

Our rhythm was simple: (1) understand the customer’s problem, (2) unblock the customer by giving them immediate support, and (3) if needed, change the product so that other customers don’t run into the same problem. We used this approach for bug fixing, of course, but also to find gaps in our product, which we would address through bigger, bolder features.

Service-oriented mindset 

One trick that I use is to think of Figma as if we were in the service industry. I imagine that Figma is a hotel or a restaurant, and our customers are our guests. It’s taking the term “Software as a Service” (SaaS)* and breaking it down, word by word, as if we are literally providing a service to people. We just happen to provide that service by writing software. This is a service-oriented mindset.

The importance of feel

When you talk to designers about their preferred tools, they start talking about how things “feel.” It’s almost like talking to a musician about their favorite instrument. 

It’s my belief that the feel of software is vital to adoption, especially when you are going for product-led growth. People respond instinctively to how something feels, and you can’t fake it. In fact, I think it might be the most important part of PLG.

Cutting multiplayer out of the initial release was really tough, but our thinking was that if we had started with something that felt bad—even with all the right features—we would’ve started with a credibility problem. We would’ve had to then spend the majority of our time convincing people that we weren’t bad anymore. Even though we didn’t have every feature from the start, this trust allowed our customers to believe that we would continue to improve and add features based on their feedback.

The elements of feel-good software

Every product will have different aspects that are important to get right for the feel of it.

For Figma, we focused on the following:

  • Performance—We measure how fast we are able to display changes as the user edits the document. We try to hold ourselves to a bar of 60 frames per second whenever we can.

  • Familiarity—Design tools have a long history of keyboard shortcuts and muscle memory. For example, certain modifier keys affect how objects get resized, and there is certain behavior that users expect if the keys are pressed or released halfway through. There are even combinations you can do, like holding down the spacebar while moving an object, that feel like advanced video game combo moves.

  • Consistency—Figma can be quite complex. Objects can be duplicated from file to file, nested within each other, and updated in real time while multiple people edit at once. We have an internal set of rules for how everything operates, and these rules help people feel like the tool is predictable, even when they can’t quite put their finger on what those rules are.

Product lessons from threads

  • Thoughtfully solving the cold start problem: Threads got everyone on Instagram’s social graph, feed algorithm,  followers and even pre-seeded content from influencers before it went live for everyone.

  • Timing matters: Meta actually pulled forward the launch of Threads after some more mishaps from Twitter, where they decided to rate limit users on Twitter and also stop allowing non-logged-in users access to tweets on Twitter.

  • Distribution is king: Threads hit 100M users today, doing so in approximately 5 days. This surpasses ChatGPT, which accomplished the same in 2 months.

Calendar as a time machine

  • Time is a curious thing. It’s a constantly flowing stream that can’t be paused, stopped, or repeated. We experience it, but we can’t control it. We can’t even touch or feel it.

  • Calendars, on the other hand, cover the entire spectrum of time. Past, present and future. They are the closest thing we have to a time machine. 

  • Calendars allow us to travel forward in time and see the future. More importantly, they allow us to change the future.

  • Changing the future means dedicating time to things that matter. It means allocating our most precious resources to activities with the highest expected return on investment.

  • Treating todos as calendar events is helpful because calendars introduce constraints. A calendar forces you to estimate how long each task will take and then find empty space for it on a 24 hours × 7 days grid, which is already cluttered with other things. It’s like playing Tetris with blocks of time.

  • We only use our calendars to look forward in time, never to reflect on things that happened in the past. We can use calendars to reflect.

Others

Originally on Substack